sign in sign up
<<
Home , Abatia, Acalypha australis, Achariaceae, Afrostyrax, Anthodiscus, Apinagia

Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch

Year: 2013

Advances in the floral structural characterization of the major subclades of , one of the largest orders of flowering

Endress, Peter K ; Davis, Charles C ; Matthews, Merran L

DOI: https://doi.org/10.1093/aob/mct056

Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-78190 Journal Article Published Version

Originally published at: Endress, Peter K; Davis, Charles C; Matthews, Merran L (2013). Advances in the floral structural characterization of the major subclades of Malpighiales, one of the largest orders of flowering plants. Annals of Botany, 111(5):969-985. DOI: https://doi.org/10.1093/aob/mct056 Annals of Botany 111: 969–985, 2013 doi:10.1093/aob/mct056, available online at www.aob.oxfordjournals.org

Advances in the floral structural characterization of the major subclades of Malpighiales, one of the largest orders of flowering plants

Peter K. Endress1,*, Charles C. Davis2 and Merran L. Matthews1 1Institute of Systematic Botany, University of Zurich, Zollikerstrasse 107, 8008 Zurich, Switzerland and 2Department of Organismic and Evolutionary Biology, Harvard University Herbaria, 22 Divinity Avenue, Cambridge, MA 02138, USA * For correspondence. E-mail [email protected]

Received: 28 November 2012 Revision requested: 21 December 2012 Accepted: 24 January 2013 Published electronically: 13 March 2013

† Background and Aims Malpighiales are one of the largest angiosperm orders and have undergone radical sys- tematic restructuring based on molecular phylogenetic studies. The has been recalcitrant to molecular Downloaded from phylogenetic reconstruction, but has become much more resolved at the suprafamilial level. It now contains so many newly identified that there is an urgent need for comparative studies to understand their structure, biology and . This is especially true because the contains a disproportionally large diversity of rain and includes numerous agriculturally important plants. This study is a first broad systematic step in this endeavour. It focuses on a comparative structural overview of the flowers across all recently identified supra-

familial clades of Malpighiales, and points towards areas that desperately need attention. http://aob.oxfordjournals.org/ † Methods The phylogenetic comparative analysis of floral structure for the order is based on our previously pub- lished studies on four suprafamilial clades of Malpighiales, including also four related rosid orders (, , , ). In addition, the results are compiled from a survey of over 3000 pub- lications on macrosystematics, floral structure and embryology across all orders of the core . † Key Results Most new suprafamilial clades within Malpighiales are well supported by floral structural features. Inner morphological structures of the (i.e. stigmatic lobes, inner shape of the , , presence of obturators) and (i.e. structure of the nucellus, thickness of the integuments, presence of vas- cular bundles in the integuments, presence of an endothelium in the inner integument) appear to be especially

suitable for characterizing suprafamilial clades within Malpighiales. at Universitaet Zuerich on April 20, 2013 † Conclusions Although the current phylogenetic reconstruction of Malpighiales is much improved compared with earlier versions, it is incomplete, and further focused phylogenetic and morphological studies are needed. Once all major subclades of Malpighiales are elucidated, more in-depth studies on promising structural features can be conducted. In addition, once the phylogenetic of Malpighiales, including closely related orders, is more fully resolved, character optimization studies will be possible to reconstruct evolution of structural and biological features within the order.

Key words: Androecium, Celastrales, COM clade, core eudicots, evolution, floral structure, gynoecium, Malpighiales, ovules, Oxalidales, phylogeny, .

INTRODUCTION include Samydaceae and Scyphostegiaceae; Passifloraceae to also include and Malesherbiaceae), and new The eudicot order Malpighiales contains approx. 40 families suprafamilial clades of families have been discovered (e.g. and includes more than 16 000 species, making it one of the the parietal clade, the rhizophoroid clade and largest orders of the flowering plants (APG III, 2009; Rafflesiaceae + ; Schwarzbach and Ricklefs, Wurdack and Davis, 2009; Xi et al., 2012). The sheer - 2000; Zhang and Simmons, 2006; Alford, 2007; Davis et al., logial diversity in the group, including such extremes as trop- 2007; Wurdack and Davis, 2009; Xi et al., 2012). Despite ical holoparasites with giant flowers and temperate and the extreme difficulty in resolving the order (Davis et al., herbs with tiny, simple flowers, has long made its classification 2005; Korotkova et al., 2009; Wurdack and Davis, 2009), challenging. First taking shape in the early 1990s (Chase et al., however, recent attempts using phylogenomic approaches 1993; APG I, 1998; Nandi et al., 1998), Malpighiales have have made significant progress, although some relationships undergone massive restructuring over the past 15 years as a remain unresolved (Wurdack and Davis, 2009; Xi et al., result of the molecular phylogenetic revolution (Savolainen 2012). In the most recent study by Xi et al. (2012) the 16 pre- et al., 2000a, b; Soltis et al., 2000; Davis and Chase, 2004; viously identified clades of Malpighiales, eight of which are APG III, 2009; Wurdack and Davis, 2009; Xi et al., 2012). suprafamilial (Wurdack and Davis, 2009), have been resolved A number of new families have been recognized (e.g. into three large clades. Nine additional suprafamilial subclades , ), others excluded (e.g. have also been resolved within these three major clades. A new Medusandraceae, ), some traditional families phylogenetic analysis of Malpighiales based on structural fea- have been alternatively circumscribed (e.g. to also tures at a more refined level is our ultimate goal but will only

# The Author 2013. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For Permissions, please email: [email protected] 970 Endress et al. — Floral structure and systematics in Malpighiales be possible to achieve once all families or suprafamilial clades appropriate selection to elucidate putative synapo- have been equally studied. This endeavour will be of interest morphies for major Malpighiales subclades, and second, as especially for the evaluation of Malpighiales, for we outline below in greater detail, numerous groups within which DNA is unavailable. In combination with molecular Malpighiales still lack sampling for morphological and ana- analyses it may also improve the phylogenetic resolution of tomical features. To rectify this, the laboratory of P.K.E. has the group as has been shown in other studies (Nandi et al., been conducting a series of comparative floral structural 1998; Doyle and Endress, 2000). studies focusing on several newly identified Malpighiales These newly recognized Malpighiales subclades provide a clades (Endress and Matthews, 2006b, 2012; Matthews and foundation for conducting a comprehensive comparative ana- Endress, 2008, 2011, 2013; Matthews et al., 2012). This re- lysis of their floral structure and biology, which to date has search programme has also included the closest relatives of been sorely lacking. The reason for this vacuum in our knowl- Malpighiales, Oxalidales and Celastrales (Matthews et al., edge is two-fold. First, the poor phylogenetic resolution (in- 2001; Matthews and Endress, 2002, 2005a), plus two addition- cluding sparse taxon sampling) within the order has hindered al rosid orders, Cucurbitales and Crossosomatales (Matthews and Endress, 2004, 2005b). The main aim of this research has been to identify special floral features that characterize

these clades based on the reconstruction of their outer and Downloaded from TABLE 1. List of families in Malpighiales (following Xi et al., inner morphological surfaces using serial microtome sections 2012) plus scanning electron microscopy, and also anatomical and histological features. Concurrently, we have conducted a vast (16/12) literature survey including over 3000 publications on macro- Balanopaceae (1/1) systematics, reproductive structure and embryology throughout

Bonnetiaceae (3/1) http://aob.oxfordjournals.org/ (4/1) all families of the core eudicots, which we have also used here. (3/0) To demonstrate the unbalanced sampling of structural fea- Centroplacaceae (0/0) tures for Malpighiales we prepared a table (Table 1) listing (9/1) the number of publications on aspects of , embryo sac (20/6) Ctenolophonaceae (5/0) and floral development as well as the inner floral structure (4/1) for each of Malpighiales [clade circumscription (6/6) Wurdack and Davis (2009) and Xi et al. (2012)]. The clades (12/8) treated in these two papers were chosen over the APG III Euphorbiaceae (172/132) Euphroniaceae (1/0) (2009) list as they provide a more extensive representation of at Universitaet Zuerich on April 20, 2013 Goupiaceae (0/0) subclades for the euphorbioids, salicoids and ochnoids. (10/2) Table 1 shows that 13 of the approx. 40 families (about 30 (13/10) % of the families) are still unknown in their ovule and (3/1) female gametophyte development, while the relatively large (3/0) (0/0) number of 132 for Euphorbiaceae is misleading because (43/17) many of these studies treat only a small number of genera Lophopyxidaceae (1/0) within the family (mainly , and ). Malesherbiaceae (1/0) In the case of the latter, the ovule and female gametophyte de- (21/12) Medusagynaceae (3/0) velopment for most of the over 200 genera of Euphorbiaceae (15/7) (Webster, 1994b) remain unknown. In addition, many of (2/0) these publications are of varying quality and comprehensive- Passifloraceae (31/13) ness, some relatively broad and others with only minor notes Peraceae (2/0) on single features from single species. The most useful publi- (19/10) (8/2) cations are those with high-quality illustrations because they (70/36) help with comparisons, especially when ambiguous terms are (5/4) used. In our summary of this work we have tried to use all (1/0) existing publications as much as possible; however, for the Rafflesiaceae (7/6) (15/6) sake of brevity we cite only a small number of these. In par- Salicaceae (26/13), ticular, we focus on the newer and more comprehensive Samydaceae (3/3) works whenever possible. Scyphostegiaceae (2/0) A prominent character emphasized in premolecular classifica- (2/1) tions of Malpighiales family groups has been placentation Turneraceae (13/5) (28/14). (Cronquist, 1981; Takhtajan, 1997). Specifically, ovaries with axile vs. parietal placentation have played a prominent early The first number indicates the number of publications covering aspects of role in macrosystematics at the order and subclass rank (see floral development and inner floral structure for members of each family Endress and Matthews, 2012). Part of the current Malpighiales, (including male and female gametophytes and palynology). The second mainly families with parietal placentation, was earlier accommo- number indicates the number of publications on ovule development and/or female gametophyte development for members of each family (the second dated in the order (or ) [including Salicaceae, number is also included in the first number because the two topics often Samydaceae, Scyphostegiaceae, Lacistemataceae, Achariaceae, overlap). Violaceae, Turneraceae, Malesherbiaceae, Passifloraceae (of clade Endress et al. — Floral structure and systematics in Malpighiales 971

1inXi et al., 2012), and Elatinaceae (clade 3)] and another level and (2) the phylogenetic structure within the clade of part, in Guttiferales [including Medusagynaceae, Ochnaceae, interest. It is also good to have an idea of the degree of stability Quiinaceae, , Clusiaceae, Calophyllaceae, Hyperi- of the selected floral features within the clade of interest at the caceae (clade 2) and Caryocaraceae (clade 3)], which included outset of the study for the purpose of identifying putative syn- mainly families with axile placentation (e.g. Melchior, 1964). apomorphies. If we study a larger group more closely some Similarly, in Cronquist (1981), in the parietal Violales [in- features remain relatively constant (e.g. details of ovule struc- cluding Salicaceae (as ), Lacistemataceae, ture in orders or supraordinal clades of angiosperms), whereas Scyphostegiaceae, Violaceae, Turneraceae, Malesherbiaceae, others fluctuate at low systematic levels (e.g. colour of flowers; Passifloraceae and Achariaceae (clade 1)] and the axile ‘Brownian Motion’, Endress, 1994). This becomes ever more [including Ochnaceae, Quiinaceae, Medusagynaceae, obvious by molecular phylogenetic studies with broad taxon Clusiaceae, Hypericaceae, Calophyllaceae and Bonnetiaceae sampling (e.g. Losos, 2011). Certain features are more prone (clade 2), and Caryocaraceae and Elatinaceae (clade 3)]. to occur (and change) in one clade but not in another, These different placental forms are still of interest and appear perhaps due to the morphological and genetic preconditions to delineate some recently defined clades (e.g. the euphorbioids for their appearance. In individual lineages they can disappear with axile placentation, and the parietal clade with parietal pla- and reappear again, such that there appears to be rampant par-

centation) but are most appropriate at lower systematic levels allel evolution (Endress, 2010, 2011b), which may also apply Downloaded from than order. at the genetic level (W.-H. Zhang et al., 2012). In our assess- With the advent of our more highly resolved phylogenetic ment here, we focus especially on features that are more tree of Malpighiales (Xi et al., 2012), it is appropriate to common in Malpighiales than in other orders, and features present an account on the current state of knowledge of the re- that show a pattern of distribution within Malpighiales that productive structures in the newly recognized suprafamilial may characterize larger or smaller subclades. Where the phylo- clades of Malpighiales. Here, we combine our previous mor- genetic tree is not resolved for all subclades, our morphologic- http://aob.oxfordjournals.org/ phological studies within Malpighiales and related groups al summary also serves as a guide to possibly unite clades that (Endress and Matthews, 2006b; Matthews and Endress, may be identified in the future. At present, given the uncertain- 2006, 2008, 2011, 2013; Matthews et al., 2012) with data ties explained in the following two paragraphs, we often available in the literature to produce a list of features and cannot determine synapomorphies or plesiomorphies for sub- their distribution across the order. Because our database clades, but we can recognize special features that are character- spans the entire core eudicots we also recognize features that istic for groups and thus hypothesize evolutionary trends are especially common in Malpighiales (i.e. occurring in within subclades.

many families and in a number of suprafamilial clades) com- With respect to the phylogenetic position of Malpighiales, at Universitaet Zuerich on April 20, 2013 pared with other orders, plus features that may be unique for our current knowledge suggests that within the rosid clade, the order. The latter are especially important as morphological Oxalidales + represent the of characters that convincingly define the order have to-date been Malpighiales (Wurdack and Davis, 2009; see also Zhang and elusive. Gynoecial (including ovular) features appear especial- Simmons, 2006; Davis et al., 2007), with Celastrales sister ly promising in this regard (for other non-nucleotide features, to Malpighiales + (Oxalidales + Huaceae) (Wurdack and see also Nandi et al., 1998; Stevens, 2001 onwards). The gy- Davis, 2009). Unfortunately, the small family Huaceae (Hua noecium is especially rich in this regard because it exhibits and ) is poorly known in its floral structure (Baas, more structural complexity than the and androecium 1972; Bayer, 2007). In contrast, the floral structure of by having an inner morphological surface and by also contain- Oxalidales and Celastrales has been investigated in detail ing the ovules. Our work is a first major step to identify floral (Matthews and Endress, 2002, 2005a). These comparative morphological features associated with the many newly iden- studies are useful for a tentative evaluation of evolutionary tified clades within the order. Here, these serve as hypothetical trends in Malpighiales. The COM clade (coined by Endress synapomorphies. Obtaining a comprehensive picture of evolu- and Matthews, 2006b, and consisting of Celastrales, tionary changes of flowers within the major subclades of Oxalidales and Malpighiales) has been shown to be more Malpighiales is a multi-step procedure, which we hope to con- closely related to malvids based on floral structure (Endress tinue to pursue in the near future, but which is beyond the and Matthews, 2006b) than to fabids, to which it was original- scope of our present study. This goal is a vast undertaking ly positioned based largely on plastid DNA phylogenies because morphological analyses are time-consuming, plus it (Savolainen et al., 2000a; Hilu et al., 2003). Several recent is more difficult to attain the necessary material than for mo- molecular phylogenetic studies, however, especially those lecular studies, for which any of the can be using mitochondrial and nuclear , indicate that the used. Moreover, key features of the phylogeny remain unre- COM clade is sister to malvids (Zhu et al., 2007; Finet solved (especially at the base), and numerous families are et al., 2010; Qiu et al., 2010; Shulaev et al., 2010; Lee poorly sampled taxonomically in terms of molecular phyloge- et al., 2011;N.Zhang et al., 2012), or nested in malvids netics (e.g. Ochnaceae). Finally, in addition to presenting our (Morton, 2011). These analyses suggest the possibility of a current state of knowledge, our summary also serves to iden- large-scale genomic incongruence involving the plastid place- tify taxa for which focused floral morphological sampling is ment of the COM clade, with more consistent desperately needed. with the mitochondrial and nuclear phylogenies. For a reconstruction of the evolution of floral features in With respect to the phylogenetic structure within Malpighiales, or any such group for that matter, we need to es- Malpighiales, Xi et al. (2012) found that the families of tablish (1) the position of the clade at a deeper phylogenetic Malpighiales form three major subclades (clades 1–3) 972 Endress et al. — Floral structure and systematics in Malpighiales

(Fig. 1); however, relationships between them are not resolved. Erythroxylaceae (2) (Ganders, 1979). Cleistogamous flowers Within these three clades are a number of lesser inclusive sub- occur in some Violaceae (1), Podostemaceae (2) and clades, each with differing molecular phylogenetic support. Elatinaceae + Malpighiaceae (3). Unisexual flowers are Within clade 1, euphorbioids are weakly supported, but sub- common in the euphorbioids (including Rafflesiaceae), and clades within are strongly supported and include Salicaceae + Achariaceae p.p., are rare in Violaceae (1), and Euphorbiaceae, the phyllanthoids and the linoids (informal also occur in Clusiaceae, Medusagynaceae + Quiinaceae (2) names following Xi et al., 2012). Mixed support is present and in putranjivoids (3). An abscission zone below the flowers in the subclade containing the parietal clade, salicoids and is common in Malpighiales (Stevens, 2001 onwards), but is related families. Clade 2 contains the clusioids, ochnoids, rhi- lacking or poorly developed in putranjivoids (Matthews and zophoroids and pandoids, and all but the pandoids are well Endress, 2013). supported as subclades, although the resolution between them is poor. In clade 3, which contains the chrysobalanoids, Perianth malpighioids, putranjivoids and Caryocaraceae, all but the malpighioids form well-supported groups. Synsepaly is scattered over the entire order but is never pre- Using this molecular plus our present dominant in families or suprafamilial groups, e.g.

knowledge of the floral structure within these groups, we Euphorbiaceae (Prenner et al., 2008a, b; De-Paula et al., Downloaded from present the first broad synthesis of the potential floral features 2011) + Peraceae (Bigio and Secco, 2012) (1); short synsepaly characterizing these subclades. In the first section we discuss is present in Humiriaceae (Winkler, 1931), some Achariaceae floral structural features of interest that occur in the various (Sleumer, 1954) + Salicaceae (Kaul, 1995) (1), in all four families. In the second section we list features that potentially families of Chrysobalanaceae sensu lato (s.l.) (except characterize the three main Malpighiales clades and the Balanopaceae) (Merino Sutter and Endress, 2003; Matthews various subclades sensu Xi et al. (2012). and Endress, 2008), rhizophoroids (Matthews and Endress, http://aob.oxfordjournals.org/ 2011), ochnoids (Matthews et al., 2012), Pandaceae (Mildbraed, 1931) (2), Elatinaceae (Niedenzu, 1925) and MORPHOLOGY OF AND THEIR Putranjivaceae (Matthews and Endress, 2013) (3). Lability PARTS may be expressed in that the male flowers of a species are syn- In this section the focus is on flowers, their parts and the pos- sepalous, but the female flowers are not (e.g. ition of specific floral features. The numbers given in paren- mcvaughii; Steinmann, 2005). Secretory structures (glands) theses (1–3) after family names or groups of families refer on the outer surface of are present in some

to the numbers of the three major subclades of Malpighiales Euphorbiaceae (Radcliffe-Smith, 2001), Linaceae (Narayana at Universitaet Zuerich on April 20, 2013 to which they belong (sensu Xi et al., 2012; Fig. 1). Closely and Rao, 1969b), Passifloraceae (Knapp and Mallet, 1984) related families within each major subclade are connected (1), Humiriaceae (2) (Winkler, 1931; Sabatier, 2004), most with a + sign (but they do not necessarily represent sister Malpighiaceae (e.g. Vogel, 1974; Anderson, 2007) + some families). Elatinaceae (Niedenzu, 1925), Caryocaraceae (Matthews and Endress, 2011), and some Chrysobalanaceae (Prance and White, 1988) (3). Secretory fimbriae on margins are Flowers known from some Euphorbiaceae (1) (Dehgan and Webster, Monosymmetric flowers occur especially in Violaceae (1), 1979), Ochnaceae (2) (Matthews et al., 2012), some chrysobalanoids and Malpighiaceae (3) (oblique in the Malpighiaceae (Anderson, 2006) and Putranjivaceae second two clades; Vogel, 1990; Matthews and Endress, (Matthews and Endress, 2013) (3). Sepals of unequal size 2008; W.-H. Zhang et al., 2012), otherwise only occasionally (outer two or three shorter than inner ones) are common in in predominantly polysymmetric families (Endress, 2012). several groups, including the ochnoids (2) (Matthews et al., Trimerous flowers are predominant in some Achariaceae + 2012), chrysobalanoids (Matthews and Endress, 2008) and Salicaceae, part of Euphorbiaceae + Phyllanthaceae + putranjivoids (Matthews and Endress, 2013) (3). Special muci- Picrodendraceae (1), part of Podostemaceae (2), and in lage cells in the sepals are common in Malpighiales as in other Elatinaceae (3). A long androgynophore occurs in some rosids (Matthews and Endress, 2006) but are notably absent in Passifloraceae (Bernhard, 1999b) + Malesherbiaceae (Harms, the putranjivoids (Matthews and Endress, 2013). 1925a), and among euphorbioids (male flowers with reduced gy- are larger than sepals in floral and cover the inner noecium) in Peraceae (; Pax and Hoffmann, 1931) + floral organs (sometimes combined with valvate aestivation) in Euphorbiaceae (; Michaelis, 1924) + Phyllantha- several families, including Humiriaceae (Winkler, 1931), ceae (; Airy Shaw, 1974) (1). A floral cup is Achariaceae (Bernhard, 1999a), Goupiaceae (Mitchell, 2002) common and occurs in Passifloraceae (Bernhard, 1999b) + and Linaceae (Matthews and Endress, 2011) (1), Turneraceae (Gonza´lez, 1993), occasionally in Salicaceae and Erythroxylaceae + Ctenolophonaceae, Irvingiaceae (Matthews Phyllanthaceae (1), in some Quiinaceae (2) (Matthews et al., and Endress, 2011), ochnoids (Matthews et al., 2012) (2), 2012), in some Rhizophoraceae + Erythroxylaceae (Matthews some chrysobalanoids (Matthews and Endress, 2008) and and Endress, 2011), and at least partly in all families of the chry- Caryocaraceae (Matthews and Endress, 2011) (3). aestiv- sobalanoids (except Balanopaceae) (Matthews and Endress, ation is contort in Humiriaceae (Narayana and Rao, 1969a), 2008) (3). Heterostyly is known from Linaceae (Matthews and Turneraceae (Gonza´lez, 1993), and Linaceae (Matthews and Endress, 2011) and Turneraceae (1) (Arbo, 2007), as well as Endress, 2011) + Ixonanthaceae (Steyermark and Luteyn, Hypericaceae (Lloyd and Webb, 1992; Robson, 2012) and 1980) (1), many clusioids + ochnoids (Amaral, 1991; Endress et al. — Floral structure and systematics in Malpighiales 973

12 345678910 4B Euphorbiaceae 4A Euphorbioids Peraceae1 ?????? Phyllanthoids 4 5A Picrodendraceae Phyllanthaceae 5 Linoids Linaceae 6 lxonanthaceae 8E Salicaceae 8D Scyphostegiaceae 8C Samydaceae 1 Salicoids Lacistemataceae ?????? 8 2 8B Downloaded from 2 7D 8A Turneraceae Parietal Malesherbiaceae2 clade 7B Violaceae 9 Goupiaceae ?????? 7C 7 Achariaceae http://aob.oxfordjournals.org/ 7A Humiriaceae 10F Hypericaceae 10E Clusioids Podostemaceae 10C Calophyllaceae 10D Clusiaceae Bonnetiaceae Malpighiales 10

Ochnoids Ochnaceae3 at Universitaet Zuerich on April 20, 2013 10A 10B Quiinaceae3 3 2 Medusagynaceae Rhizophoroids 2B Rhizophoraceae 2A Erythroxylaceae Ctenolophonaceae Pandoids Pandaceae ???? 11 lrvingiaceae 3D Chrysobalanaceae 3C Euphroniaceae Chrysobalanoids 3A 3E Dichapetalaceae Trigoniaceae 3B Balanopaceae Malpighioids 12A Malpighiaceae Elatinaceae 3 12 Centroplacaceae ?????? 3F Caryocaraceae Putranjivoids 3G Putranjivaceae Lophopyxidaceae

F IG. 1. Phylogeny of Malpighiales reproduced from Xi et al. (2012) (Peraceae and Rafflesiaceae follow Wurdack and Davis, 2009). Suprafamilial clades are discussed in the main text (numbers 1–12 follow Xi et al., 2012). The following families are treated differently in APG III (2009): (1) Peraceae are circumscribed with Euphorbiaceae; (2) Passifloraceae, Turneraceae and Malesherbiaceae are circumscribed as Passifloraceae s.l.; (3) Ochnaceae, Medusagynaceae, and Quiinaceae are circumscribed as Ochnaceae s.l. The distribution of ten prominent features of gynoecium (incl. ovule) structure is plotted: 1. Flowers with mainly 3 carpels. 2. Carpels with 2 ovules, collateral, antitropous, with obturator. 3. Carpels with 1 ovule (median). 4. Carpels with more than 2 ovules. 5. Ovules crassinucellar. 6. Ovules with thin nucellus (incompletely tenuinucellar or weakly crassinucellar). 7. Inner integument thicker than outer at the mature embryo sac stage. 8. Outer integument thicker than inner at the mature embryo sac stage. 9. Outer integument only 2 cell layers thick at the mature embryo sac stage. 10. Inner integument 5 or more cell layers thick at the mature embryo sac stage 974 Endress et al. — Floral structure and systematics in Malpighiales

Matthews et al., 2012), some Rhizophoraceae + 400; Stevens, 1976) + Hypericaceae (650; Robson, 1996) Ctenolophonaceae (Matthews and Endress, 2011) (2), and (2). united in fascicles (or phalanges) occur especially Euphroniaceae + Trigoniaceae (3) (Matthews and Endress, in the clusioids (2), such as in Clusiaceae (Leins and Erbar, 2008). Petals with an often conspicuously narrow base are 1991; Bittrich and Amaral, 1996; Stevens, 2007; Sweeney, present in Linaceae (1) (Matthews and Endress, 2011), 2008), and Hypericaceae (Leins, 1964; Baas, 1970; Ronse Rhizophoraceae + Erythroxylaceae (Matthews and Endress, Decraene and Smets, 1991) + Podostemaceae (a single 2011), many taxa in the three families of ochnoids (Matthews phalanx) (Rutishauser, 1997). One-sided (monosymmetric) et al., 2012) (2), and in Chrysobalanaceae s.l. (Matthews and union of groups of stamens is prominent in Endress, 2008). Petals with marginal elaborations are most pro- Chrysobalanaceae s.l. (Prance and White, 1988; Matthews nounced in some Linaceae (1), many Rhizophoraceae (2) and Endress, 2008) (3). Synandry of all stamens (at least for (Matthews and Endress, 2011) and Malpighiaceae (3) (Davis a short distance) is known from many clades, e.g. some and Anderson, 2010). Petals with ventral elaborations occur in euphorbioids (1), rhizophoroids, few Ochnaceae (2), most Linaceae (1) and some Erythroxylaceae (2) (Endress and chrysobalanoids, Caryocaraceae and few Putranjivaceae (3) Matthews, 2006a; Matthews and Endress, 2008). (Matthews and Endress, 2008, 2011, 2013; Matthews et al., Petals are postgenitally connected in various ways in some 2012). Reduction to a single has repeatedly occured

Linaceae (Matthews and Endress, 2011) and of and is known from several families, including Euphorbiaceae Downloaded from Euphorbiaceae (Kapil, 1956) (1), and some (Prenner and Rudall, 2007), Lacistemataceae (Agostini, Rhizophoraceae + Ctenolophonaceae (2) (Matthews and 1973; Endress, 1999) (1), Ochnaceae (Matthews et al., 2012) Endress, 2011). Sympetaly has been mentioned for some and Podostemaceae (Thiv et al., 2009) (2). Euphorbiaceae (Pax and Hoffmann, 1931); however, it is not There are relatively few floral developmental studies in established whether this is real sympetaly (i.e. congenital Malpighiales, and thus broad comparisons of androecium de- union) or only postgenital union, and whether the perianth velopment are not yet possible. However, in polystemonous http://aob.oxfordjournals.org/ organs in these cases are even homologous with petals. Salicaceae, stamen initiation is centrifugal, while it is centri- The unusual condition of three (instead of one) vascular petal or more or less synchronous in Achariaceae (Bernhard traces in petals occurs in Caryocaraceae (3) and Irvingiaceae and Endress, 1999; both families formerly included in (2) (Matthews and Endress, 2011), and also in some taxa of Flacourtiaceae, Chase et al., 2002) (1). Centrifugal stamen ini- all three families of ochnoids (2) (Matthews et al., 2012). tiation is also present in the clusioid families Clusiaceae A corona (an elaboration between the corolla and the an- (Hochwallner and Weber, 2006), and Hypericaceae (Leins, droecium) is known from part of the parietal clade (1) 1964) + Podostemaceae (Rutishauser and Grubert, 1999) (2),

(Malesherbiaceae, Passifloraceae, Turneraceae, Salicaceae; but in Ochnaceae (2) both centrifugal (; Amaral at Universitaet Zuerich on April 20, 2013 Bernhard, 1999b; Arbo, 2007; Kubitzki, 2007a, b; and Bittrich, 1998) and centripetal (, Pauze´ and Hemingway et al., 2011), from the clusioid-ochnoid clade Sattler, 1978) initiation is present. In Euphorbiaceae there is (Clusiaceae, Gustafsson, 2000, and Ochnaceae, Amaral, centripetal stamen initiation in (Ronse Decraene and 1991; Matthews et al., 2012), and from the rhizophoroids Smets, 1992) but probable centrifugal fascicle initiation in (Ctenolophonaceae, Matthews and Endress, 2011) (2). (Euphorbiaceae; Prenner et al., 2008b) (1). In all chrysobalanoid families (except Balanopaceae) the anthers are extremely introrse, with the thecae nearly in one Androecium plane, and the anthers have a dorsal pit (Matthews and Constantly low stamen numbers (diplostemony or haploste- Endress, 2008) (3). To have the thecae in one plane is also mony) are found in Linaceae, the parietal clade (most characteristic of Podostemaceae (Kapil, 1970) (2). Passifloraceae, Turneraceae + Malesherbiaceae, Violaceae + Goupiaceae, Lacistemataceae, and some Salicaceae) (1), pan- Gynoecium doids (Pandaceae + Irvingiaceae) (2) and malpighioids (Malpighiaceae + Elatinaceae) (3). Haplostemony is especial- Trimery is most common in the gynoecium of families ly present in part of the parietal clade (Malesherbiaceae, within Malpighiales and is especially prominent in the euphor- Passifloraceae + Turneraceae, and Violaceae + Goupiaceae) bioids and the parietal clade (1), and in the malpighioids (3). (1). However, there are multiple trends to polystemony (e.g. An increase of carpel number to more than five has occurred from haplostemony in Passifloraceae; Bernhard, 1999b; numerous times in various families without any obvious con- Krosnick et al., 2006). Stamen numbers with broad ranges centration in a single subclade. Also, pseudomonomery (gy- occur in several larger groups, such as Humiriaceae, noecia with a single fertile carpel and reduced sterile Achariaceae, Salicaceae and Euphorbiaceae (1), carpels) is known from several distantly related clades, includ- Clusiaceae + Podostemaceae (2), and Putranjivaceae (3). ing Euphorbiaceae (1), Chrysobalanaceae and Putranjivaceae Representatives with especially large stamen numbers (up to (3) (Matthews and Endress, 2008, 2013). several hundred) include Humiriaceae (200; Sabatier, 2002), Angiospermy type 4 (for typology, see Endress and Salicaceae (400; Alford, 2008), Euphorbiaceae (up to 1000 Igersheim, 2000) was found in all chrysobalanoids studied in Ricinus; Radcliffe-Smith, 2001) (1), Caryocaraceae (750; (3) (Matthews and Endress, 2008) and in most rhizophoroids Prance and da Silva, 1973), Chrysobalanaceae (300; Prance (2) (Matthews and Endress, 2011), whereas type 2 is predom- and White, 1988), some ochnoids + clusioids, such as inant in putranjivoids (3) (Matthews and Endress, 2013). In Medusagynaceae (480; Matthews et al., 2012) + Quiinaceae ochnoids (2) types 2, 3 and 4 occur (Matthews et al., 2012). (600; Matthews et al., 2012), and Calophyllaceae (more than Postgenital union of free carpel parts appears to be rare; it Endress et al. — Floral structure and systematics in Malpighiales 975 was found in two distantly related families: Trigoniaceae (3) carpel is a unique syndrome in the COM clade among eudi- (Trigoniastrum; Matthews and Endress, 2008) and cots. This syndrome is most prominent in Malpighiales and Ochnaceae (2) (Ochnoideae; Matthews et al., 2012). In less so in Celastrales (, Matthews and groups that have free carpel tips they tend to be unifacial, es- Endress, 2005a) and Oxalidales (Brunelliaceae, Matthews pecially in Linaceae (Baum-Leinfellner, 1953; Matthews and and Endress, 2002). Among Malpighiales it occurs in several Endress, 2011) and Passifloraceae (Bernhard, 1999b) (1), the larger subclades, notably in the phyllanthoids (incl. rhizophoroid families Ctenolophonaceae (Matthews and Linaceae) (1), rhizophoroids (2), chrysobalanoids and putran- Endress, 2011) + Erythroxylaceae (Baum-Leinfellner, 1953; jivoids (3). Among these groups the phyllanthoids stand out Matthews and Endress, 2011), perhaps Podostemaceae from the rest due to their crassinucellar ovules (Tokuoka and (Ameka et al., 2002; Sobral-Leite et al., 2011) (2) and Tobe, 2001; Merino Sutter et al., 2006), often with nucellar Caryocaraceae (Matthews and Endress, 2011) (3). Bulged caps (and sometimes nucellar beaks), whereas the other ovaries occur in taxa of Euphorbiaceae (Pax and Hoffmann, clades have ovules with thin (slender) nucelli, mostly incom- 1931) + Picrodendraceae (Berg, 1975) (1), Ochnaceae pletely tenuinucellar or weakly crassinucellar (Endress, (Matthews et al., 2012), and Podostemaceae (Ja¨ger-Zu¨rn, 2011a), commonly with an endothelium and early disintegra- 2003) (2), Caryocaraceae (Matthews and Endress, 2011), tion of the nucellus around the embryo sac (Boesewinkel

Chrysobalanaceae (Juel, 1915; Prance and White, 1988; and Bouman, 1980; Tobe and Raven, 1984; Matthews and Downloaded from Matthews and Endress, 2008) and Elatinaceae (Niedenzu, Endress, 2008, 2011, 2013). In Centroplacaceae () and 1925) + Malpighiaceae (Simpson, 1989) (3). Balanopaceae (3) the two ovules are not pendant but upright Parietal placentae are predominant in the ‘parietal clade’, in- (Loesener, 1942; Merino Sutter and Endress, 2003). A rarer cluding the families Achariaceae, Goupiaceae + Violaceae, variant is carpels with more or less collateral placentae but Malesherbiaceae, Turneraceae + Passifloraceae, Lacistemata- the ovules are curved in a way that they are positioned one ceae, Samydaceae, and Salicaceae + Scyphostegiaceae (1). upon the other, such as in Humiriaceae (1) (Cuatrecasas, http://aob.oxfordjournals.org/ Parietal placentae also tend to go hand-in-hand with several 1961; Boesewinkel, 1985), the ochnoids Quiinaceae + (or at least more than two) ovules per carpel. More than two Medusagynaceae (2) (Matthews et al., 2012), and in the chry- ovules per carpel are predominant in the parietal clade (1) and sobalanoids Balanopaceae + Euphroniaceae (3) (Merino in clusioids (2) and are present in only few families of other Sutter and Endress, 2003; Matthews and Endress, 2008). Malpighiales subclades. The low ovule number in the parietal Another variant is carpels with just a single ovule, which Lacistemataceae (Krause, 1925) probably arose by reduction occurs in Peraceae (Bigio and Secco, 2012) + Euphorbiaceae and is correlated with flower miniaturization. Extreme placenta- (Pax and Hoffmann, 1931), Lacistemataceae (in part; Krause,

tion forms are the free central placentae in some Podostemaceae 1925) (1), Irvingiaceae (Matthews and Endress, 2011; Tobe at Universitaet Zuerich on April 20, 2013 (Moline et al., 2007; Thiv et al., 2009), and the nearly pla- and Raven, 2011) + Pandaceae (Mildbraed, 1931; Pax and centae in Scyphostegiaceae (1) (van Heel, 1967) and some Hoffmann, 1931), Ochnoideae of Ochnaceae (Matthews species of Ochnaceae (2) (Sastre, 1970). A notable et al., 2012) (2), Caryocaraceae (Dickison, 1990a; Matthews trend is the advent of false septa developing from the dorsal and Endress, 2011) and Malpighiaceae (Singh, 1959; Souto part of each carpel [e.g. Linaceae (Matthews & Endress, and Oliveira, 2008) (3). In Euphorbiaceae with their uniovulate 2011), and weakly (incipient) in Ixonanthaceae (Link, 1992), carpels, the obturator develops as a double structure Picrodendraceae (Merino Sutter et al., 2006)] (1), some (Schweiger, 1905), as in the biovulate euphorbioids. It would Rhizophoraceae (Matthews and Endress, 2011), Hypericaceae be of interest to know whether there are obturators and antitro- (Eliaea, Baas, 1970) + some Calophyllaceae (Stevens, 2007) pous ovule curvature in the other uniovulate families. As in (2), and Chrysobalanaceae (Juel, 1915; Matthews and Endress, phyllanthoids the ovules in Euphorbiaceae are crassinucellar 2008) (3). and often have nucellar beaks (Bor and Bouman, 1975; Bifurcated stigmatic branches are especially known from Tokuoka and Tobe, 1995, 1998, 2002, 2003). many Euphorbiaceae (incl. Peraceae) + Phyllanthaceae (1). It Ovules with slender nucelli (incompletely tenuinucellar or is of interest that they appear to be especially concentrated weakly crassinucellar) and an endothelium (without collateral also in some other families with reduced flowers and pol- placentation and pendant ovules) also characterize other sub- lination, including Balanopaceae (Merino Sutter and Endress, clades of Malpighiales. They are predominant in ochnoids 2003), Putranjivaceae [in the wind-pollinated (Matthews et al., 2012) and clusioids (Bonnetiaceae, Prakash (!); Matthews and Endress (2013)] (3), Salicaceae and Lau, 1976; Clusiaceae, Lim, 1984; Calophyllaceae, (Kaul, 1995) + some Achariaceae (Bernhard, 1999a) (1), Moura˜o and Beltrati, 2000; Hypericaceae, Rao, 1957, + and Podostemaceae (, also wind-pollinated; Podostemaceae, Arekal and Nagendran, 1977) (2); and they Rutishauser, 1997; Philbrick and Novelo, 2004) (2). A correl- are especially thin in the sister families Hypericaceae + ation of bifurcate stigmatic branches with wind pollination also Podostemaceae. Among euphorbioids (1) incompletely tenui- appears in other orders (, Cucurbitales, , pers. nucellar ovules occur as an exception in the parasitic obs. by P.K.E.). This feature may be selectively advantageous Rafflesiaceae, where they are quite big presumably due to for maximizing capture. their large cell size (Igersheim and Endress, 1998). The unusual combination of crassinucellar ovules with an endothe- lium is present in some Linaceae + Ixonanthaceae (1) Ovules (Narayana, 1970; Matthews and Endress, 2011) and in some The presence of two collateral pendant antitropous ovules Rhizophoraceae + Erythroxylaceae (2) (Matthews and (Endress, 2011a) with obturators and axile placentae in each Endress, 2011). Although these ovules are technically 976 Endress et al. — Floral structure and systematics in Malpighiales crassinucellar they tend to have relatively thin, slender nucelli. and Salix in Salicaceae (Nikolayeva, 1983; Steyn The presence of an endothelium (or integumentary tapetum) et al., 2004) (1), in Calophyllaceae (Moura˜o and has long been known mainly from asterid families with their Beltrati, 2000), and rarely in Ochnoideae, Ochnaceae thin nucelli (Kapil and Tiwari, 1978). Our analysis shows (Matthews et al., 2012) (2) and in that the endothelium is also an important feature of the Caryocaraceae (Dickison, 1990a, b). COM clade in rosids. Here, it is nearly absent in fabids, but Vascular bundles in the outer or inner integument (never in occurs in of malvids. This further suggests that both) are present in Euphorbiaceae (inner or outer; Tokuoka more recent analyses using nuclear and mitochondrial genes, and Tobe, 1995, 2002, 2003), Phyllanthaceae (outer; but not plastid ones, may better reflect the phylogenetic place- Tokuoka and Tobe, 2001) + Picrodendraceae (inner; Merino ment of the COM clade. Sutter et al., 2006), Salicaceae (outer; Dathan and Singh, Crassinucellar ovules are also concentrated in some larger 1979) (1), Irvingiaceae (outer; Matthews and Endress, 2011; subclades, especially in the parietal clade (Achariaceae, Tobe and Raven, 2011), Rhizophoraceae (outer; Nikiticheva Steyn et al., 2003, Violaceae, Singh, 1963, Passifloraceae, and Yakovlev, 1985; Matthews and Endress, 2011) (2) and Kloos and Bouman, 1980, Turneraceae, Kloos and Bouman, Putranjivaceae (outer; Tokuoka and Tobe, 1999; Matthews 1980, + Malesherbiaceae, Ricardi, 1967, and Samydaceae, and Endress, 2013) (3). These integumentary bundles usually

Narayanaswami and Sawhney, 1959, Scyphostegiaceae, van depart from the ovular bundle at the chalaza. However, Downloaded from Heel, 1967, + Salicaceae, Steyn et al., 2005), Humiriaceae bundles of the outer integument may already depart from the (Boesewinkel, 1985), and in euphorbioids (Euphorbiaceae, raphe, which is the case in the big-seeded mangrove genera Phyllanthaceae + Picrodendraceae, Corner, 1976; Tokuoka of Rhizophoraceae (e.g. ), in Irvingiaceae and Tobe, 1998, 2001, 2002, 2003; Merino Sutter et al., 2006; (Matthews and Endress, 2011), some Ochnoideae but not in the related linoids and Rafflesiaceae) (1), and in (Ochnaceae; Matthews et al., 2012) (2), Chrysobalanaceae Malpighiaceae (3). These euphorbioids (and also (Matthews and Endress, 2008), Caryocaraceae (Matthews http://aob.oxfordjournals.org/ Malpighiaceae) are further characterized by a nucellar beak, and Endress, 2011) and rarely in Putranjivaceae (Matthews which develops by excessive growth of the nucellar apex, and and Endress, 2013) (3). The mature embryo sac grows out of is then often not covered by a micropyle (e.g. Z.-G. Zhang the nucellus tip in Samydaceae (Narayanaswami and et al., 2012), an otherwise rare feature in eudicots. The opposite Sawhney, 1959) + Salicaceae (Mauritzon, 1936; Steyn et al., kind of excessive nucellus elongation – elongation at the base 2005), which is also found in Indorouchera of Linaceae (and lack of micropyle) – occurs in Podostemaceae (e.g. (Narayana, 1970) (1). Ja¨ger-Zu¨rn, 1967) (2), and a similar basal elongation is also Another highly unusual feature within angiosperms is the

known from Elatinaceae (Dathan and Singh, 1971) (3). In fam- -type of embryo sac development. It is almost only at Universitaet Zuerich on April 20, 2013 ilies with parietal placentation, such as Passifloraceae, the ovules known in Malpighiales across all angiosperms [apart from are syntropous (e.g. Endress, 2011a). However, from fig. 8 in Penaea in , where it was first described by Stephens Singh (1963) it appears that in Violaceae antitropous ovules (1909), and Biebersteinia of (Kamelina and do occur. Thus, it would be of interest to know whether antitro- Koonova, 1990)]. Among Malpighiales it occurs in some pous ovules are not restricted to such groups with a single ovule Euphorbiaceae (e.g. Modilewski, 1910; Tateishi, 1927; or a collateral pair of ovules per carpel, but have a wider distri- Mukherjee, 1958) (1) and Malpighiaceae (e.g. Stenar, 1937; bution in Malpighiales. Rao, 1940, 1941) (3). Another shared feature in ovules A feature that has been shown to be of great macrosyste- between Euphorbiaceae and Malpighiaceae is the (occasional) matic interest in core eudicots is the thickness of the two inte- presence of several (more than two) meiocytes in an ovule, as guments at anthesis (Endress and Matthews, 2006b). That the recorded in Euphorbia (Modilewski, 1911) and inner integument is thicker (with more cell layers) than the (Singh, 1965), as well as and (Rao, outer is practically absent in the fabids, but is predominant 1941). in malvids and the COM clade. Within Malpighiales subclades with arils of various differentiation occur predominantly there are more refined patterns. In Humiriaceae and part of the in the parietal clade (Achariaceae, Violaceae, Malesherbiaceae + parietal clade, the inner integument is likewise thicker than the Passifloraceae + Turneraceae, and Samydaceae + Scyphoste- outer, but both integuments are in general thinner than in other giaceae + Salicaceae), in Linaceae + Ixonanthaceae, and in subclades: the outer often with only two cell layers, the inner Euphorbiaceae + Picrodendraceae (1), and also in Rhizophora- with often only three (Humiriaceae, Boesewinkel, 1985, ceae + Ctenolophonaceae (2) (e.g. Corner, 1976). Passifloraceae, Turneraceae, Kloos and Bouman, 1980, + Malesherbiaceae, Ricardi, 1967), while in Violaceae both inte- Floral biology guments are three cell layers thick (Singh, 1963) (1). The clu- sioid clade (2) is an exception in having predominantly the Trends to wind pollination (including flower reduction) are outer integument thicker than the inner (Bonnetiaceae, known from several subclades of Malpighiales: salicoids Prakash and Lau, 1976, Clusiaceae, Lim, 1984, + (Sacchi and Price, 1988), euphorbioids (1), chrysobalanoids Podostemaceae, Ja¨ger-Zu¨rn, 1967), with the inner only 2–3 (Balanopaceae; Merino Sutter and Endress, 2003) and putran- cell layers thick. Also in the ochnoids (2) there is a weak ten- jivoids (Matthews and Endress, 2013) (3). Among euphor- dency towards the outer integument being thicker than the bioids the transition to wind-pollination has occurred many inner (Matthews et al., 2012). times, in Phyllanthaceae + Picrodendraceae (Webster, Almost all Malpighiales have bitegmic ovules. Unitegmic 1994a), and among Euphorbiaceae in + ovules are present only in rare, scattered cases, such as (Webster, 1994a), and perhaps several times in Endress et al. — Floral structure and systematics in Malpighiales 977 each family and . The general presence of small, uni- zigzag-shaped micropyle; inner integument more than 5 cell sexual flowers without a corolla in euphorbioids may have layers thick; lobes of outer integument especially cytoplasm- been a favourable precondition for this trend. rich; nucellus long and slender and early disintegrating; endo- A unique pollination mode is via resin collection from resin- thelium present. Nectaries on androecial tube. Seeds often producing flowers (Armbruster, 1984). Among eudicots such with aril. (Matthews and Endress, 2011). flowers are only known in Malpighiales, and here in two unre- lated subclades: in euphorbioids (1), among Euphorbiaceae (2B) Rhizophoraceae + Erythroxylaceae (100/1). Floral cup (; Armbruster, 1984), and in clusioids (2), present. Free parts of sepals postgenitally connected in bud, among Clusiaceae (, , Tovomitopsis; sepals with distinctive layer of large idioblasts in the meso- Bittrich et al., 2003) + Calophyllaceae (; Bittrich phyll. Petals with conspicuously narrow attachment zone, and Amaral, 1997). Even in clusioids they most likely valvate at least close to the base and postgenitally connected evolved more than once, as indicated by their occurrence in for some distance above the base, conduplicate and enwrap- the phylogenetic tree (Gustafsson and Bittrich, 2002; Ruhfel ping stamens or parts of stamens in bud. Antepetalous et al., 2011). Other floral biological differentiations vary at stamens longer than antesepalous ones. Carpels often 3; ten- lower systematic levels and are therefore not mentioned here. dency of carpel bulging and apical septum in the . (Matthews and Endress, 2011). Downloaded from FLORAL MORPHOLOGICAL (3) Chrysobalanoids + malpighioids + putranjivoids + Caryo- CHARACTERIZATION OF MALPIGHIALES caraceae (81/1). Tendency to (oblique) floral monosymmetry SUBCLADES in chrysobalanoids and Malpighiaceae. Tendency to bulging of ovaries; placentation mostly axile. (Vogel, 1990; In this section we focus on the suprafamilial subclades of

Matthews and Endress, 2008, 2011, 2013; W.-H. Zhang http://aob.oxfordjournals.org/ Malpighiales and how they are characterized by floral features. et al., 2012). All suprafamilial subclades sensu Xi et al. (2012) are addressed (see also Fig. 1). In addition, subclades consisting (3A) Chrysobalanoids (Balanopaceae + Trigoniaceae + Dicha- of a single family are also considered if the family has an un- petalaceae + Euphroniaceae + Chrysobalanaceae) (100/1). resolved position or if its sister consists of more than two fam- Flowers obliquely monosymmetric (not in Balanopaceae). ilies. The subclades are numbered and named as above, Sepals basally united, with special mucilage cells (not in following Xi et al. (2012). Additional subclades recognized Balanopaceae). Petals of unequal size (not in Balanopaceae). but not informally named in Xi et al. (2012) are numbered Fertile stamens united to some degree into a strap on the anter-

here with the number of the inclusive subclade and an add- ior side of the flower; anthers with basal pit, strongly introrse at Universitaet Zuerich on April 20, 2013 itional specifiying letter. Characters given for the most inclu- (with thecae nearly in one plane) (not in Balanopaceae). sive clades are not repeated for the subordinated clades, Carpels mostly 3; angiospermy type 4. Ovules with thin nucel- unless the emphasis differs. Only a few publications are lus (incompletely tenuinucellar or weakly crassinucellar); cited here. More can be found in the preceding section. The inner integument more than 5 cell layers thick (only in molecular support values from Xi et al. (2012) shown below Trigoniaceae fewer than 5 cell layers thick), with endothelium. are given in parentheses after each grouping (maximum- (Litt and Chase, 1998; Merino Sutter and Endress, 2003; likelihood bootstrap percentages/Bayesian posterior probabil- Matthews and Endress, 2008). ities). The unequal length of each description below generally reflects our unequal knowledge of these groups. (3B) Balanopaceae (support not indicated, single species sampled for terminal clade). Flowers unisexual, simple, wind-pollinated. Sepals lacking or minute. Petals lacking. Stamens with short (1) Euphorbioids + Humiriaceae + parietal clade (85/1). filaments. Gynoecium 2–3-carpellate, stigmatic branches Androgynophore. Ovules mostly crassinucellar. (Pax and multiply bifurcate, large. Ovules 2 per carpel, collateral Hoffmann, 1931; Winkler, 1931; Cuatrecasas, 1961; Kapil but curved in such a way that one is above the other, weakly and Bhatnagar, 1994; Tokuoka and Tobe, 2003, 2006). crassinucellar, notably bitegmic (not unitegmic as stated in earlier publications), both integuments equally thick. (2) Clusioids + ochnoids + rhizophoroids + pandoids (83/1). (Merino Sutter and Endress, 2003). Centrifugal stamen initiation in polystemonous flowers of clu- sioids and ochnoids. Ovules with thin nucellus (weakly crassi- (3C) Trigoniaceae + Dichapetalaceae + Euphroniaceae + Chryso- nucellar or incompletely tenuinucellar). (Matthews and balanaceae (100/1). See 3A for characteristics. (Matthews and Endress, 2008; Matthews et al., 2012). Endress, 2008). (2A) Rhizophoroids (Ctenolophonaceae + Erythroxylaceae + (3D) Euphroniaceae + Chrysobalanaceae (100/1). Floral cup Rhizophoraceae) (100/1). Sepals with fewer than 3 vascular distinctive, forming a spur. Petals widely spaced on floral traces. Stamens forming a short basal androecial tube; filament cup, distinctly clawed, with lignified hairs. Style with distinct attachment of one or both of the two stamen whorls not on the longitudinal furrow between carpels; symplicate zone present rim of the androecial tube but slightly inside the tube. Short in ovary. Nectary smooth, unlobed. Cells with oxalate crystals (andro)gynophore. Angiospermy type 4 predominant; placen- present in all floral organs. (Matthews and Endress, 2008). tation axile; ovary septum with tendency to be thin and disintegrating during development leading to a secondarily (3E) Trigoniaceae + Dichapetalaceae (100/1). Floral cup unilocular ovary. Ovules antitropous and pendant, with absent. Anthers almost basifixed. Ovary and lower style 978 Endress et al. — Floral structure and systematics in Malpighiales completely synascidiate. Ovules pendant, with obturator. (7) Humiriaceae + parietal clade (62/0.79). Flowers mostly hap- Nectary with distinct scales or lobes. (Matthews and lostemonous (polystemonous in many Humiriaceae, Endress, 2008). Achariaceae and salicoids). Carpels often 3 (not in Humiriaceae); placentation parietal (not in Humiriaceae). (3F) Caryocaraceae (100/1). Flowers large. Pronouncedly poly- Ovules often more than 2 per carpel (not in Humiriaceae), stemonous, with up to 500 and more stamens per flower. Styles crassinucellar, without endothelium. Seeds often with aril free, unifacial. A single antitropous ovule per carpel, hemitro- (not in Humiriaceae). Some families representing basal pous or slighly campylotropous at anthesis, weakly crassinu- lineages (including Humiriaceae, Goupiaceae, Violaceae) cellar, without endothelium, without obturator, tendency of have a conspicuous anther appendage. (Winkler, 1931; the outer integument to be only 2 cell layers thick. Cuatrecasas, 1961; Bernhard, 1999a, b). (Dickison, 1990a; Matthews and Endress, 2011). (7A) Humiriaceae (100/1). Diplostemonous to highly polyste- (3G) Putranjivoids (Putranjivaceae + Lophopyxidaceae) (100/1). monous. Stamens basally united into a tube, anthers with con- Flowers unisexual. Petals lacking or small. Stamens in one spicuous broad appendage. Carpels often 5, forming a united , antesepalous (or probably secondarily increased). style; placentation axile. Ovules 1 or 2 per carpel, antitropous Angiospermy type 2 predominant; placentation axile. Ovules (without obturator?), pendant, often at unequal levels (one

2 per carpel, antitropous, pendant, with obturator, with thin nu- Downloaded from upon the other); integuments relatively thin (outer 2 layers, cellus (weakly crassinucellar or incompletely tenuinucellar), inner 3 layers). (Winkler, 1931; Mauritzon, 1934; inner integument more than 5 cell layers thick. Mucilage Cuatrecasas, 1961; Boesewinkel, 1985). cells and special mucilage cells in floral organs lacking. indehiscent, . (Pax and Hoffmann, 1931; (7B) Parietal clade (Achariaceae + Violaceae + Goupiaceae + Matthews and Endress, 2006, 2013). salicoids) (100/1). Placentation parietal. (Gilg, 1925; http://aob.oxfordjournals.org/ (4) Euphorbioids (Peraceae + Rafflesiaceae + Euphorbiaceae + Bernhard, 1999a, b; Bernhard and Endress, 1999). phyllanthoids + linoids) (64/0.61). Tendency to unisexual, tri- (7C) Achariaceae (100/1). Sepals often 3. Petals often more merous flowers with reduced petals (not in Linaceae + than 5. Often polystemonous. Androecium initiation centripet- Ixonanthaceae; the giant flowers of Rafflesiaceae being differ- al or synchronous. Fruits mostly indehiscent. (Gilg, 1925; ent and difficult to compare with other euphorbioids, except Lemke, 1988; Bernhard and Endress, 1999). for unisexuality). Petals, if present, often contort. . Placentation mostly axile. Ovules 2 (more rarely 1) per (7D) Violaceae + Goupiaceae + salicoids (75/0 56). See 7B. carpel, antitropous, pendant, with obturator, vascular bundles Characterization difficult. (Melchior and Becker, 1925; present in outer or inner integument (not in Linaceae + Loesener, 1942; Bernhard, 1999a; Bernhard and Endress, at Universitaet Zuerich on April 20, 2013 Ixonanthaceae). (Sutter and Endress, 1995; Bouman and 1999). Meijer, 1994; Merino Sutter et al., 2006; Matthews and (8) Salicoids (Passifloraceae alliance + Salicaceae alliance) (96/ Endress, 2011). 1). Mostly haplostemonous (not in many Salicaceae and not in (4A) Peraceae + Rafflesiaceae + Euphorbiaceae (Rafflesiaceae many Samydaceae). Corona present in some families. Ovules not included in Xi et al., 2012, thus support not indicated here). mostly more than 2 per carpel, crassinucellar (ovule structure Flowers unisexual. Sepals lacking in some Peraceae and in Lacistemataceae unknown). Seeds often with aril. (Gilg, some Euphorbiaceae. Petals mostly lacking. Flowers with a 1925; Harms, 1925a, b; Narayanaswami and Sawhney, 1959; single ovule per carpel (these features not in Rafflesiaceae). van Heel, 1967; Bernhard, 1999a, b; Bernhard and Endress, (Pax and Hoffman, 1931; Igersheim and Endress, 1998). 1999). (4B) Euphorbiaceae + Rafflesiaceae (Rafflesiaceae not included, (8A) Passifloraceae alliance (Malesherbiaceae + Turneraceae + thus support not indicated here). Characterization difficult. Passifloraceae) (100/1). Floral cup present. Corona present. (Pax and Hofmann, 1931; Bouman and Meijer, 1994; Androgynophore present (in Malesherbiaceae and Igersheim and Endress, 1998) Passifloraceae). Integuments thin (outer 2 cell layers, inner 3 cell layers). Heterostyly in Malesherbiaceae and Turneraceae. (5) Phyllanthoids + linoids (84/1). Tendency to false septa (in (Harms, 1925a, b; Ricardi, 1967; Gonza´lez, 1993; Bernhard, addition to normal septa) in the ovary. Tendency of outer in- 1999b; Arbo, 2007; Kubitzki, 2007a, b). tegument to be only 2 cell layers thick. Fruits predominantly capsules. (Pax and Hoffmann, 1931; Sutter and Endress, (8B) Passifloraceae + Turneraceae (100/1). Flowers ephemeral 1995; Merino Sutter et al., 2006; Matthews and Endress, (open one morning). Sepals mucronate. Stigma with conspicu- 2011). ous multicellular papillae. (Endress, 1994; Bernhard, 1999b; Arbo, 2007). (5A) Phyllanthoids (Phyllanthaceae + Picrodendraceae) (100/1). Flowers with tendency to trimery. Petals, if present, not con- (8C) Salicaceae alliance (Lacistemataceae + Samydaceae + spicuous. (Pax and Hoffmann, 1931; Merino Sutter et al., Scyphostegiaceae + Salicaceae) (100/1). Petals, if present, not 2006). conspicuous. (Gilg, 1925; Narayanaswami and Sawhney, 1959; van Heel, 1967; Bernhard and Endress, 1999). (6) Linoids (Linaceae + Ixonanthaceae) (96/1). Flowers bisex- ual; mostly diplostemonous. Carpels with false septa (in add- (8D) Samydaceae + Scyphostegiaceae + Salicaceae (100/1). ition to normal septa) in the ovary; placentation axile. (Link, Petals often absent (perhaps not Scyphostegiaceae). Often 1992; Matthews and Endress, 2011). polystemonous (not Scyphostegiaceae). (Gilg, 1925; Endress et al. — Floral structure and systematics in Malpighiales 979

Narayanaswamy and Sawhney, 1959; van Heel, 1967; to the other families within this clade). (Moura˜o and Beltrati, Bernhard and Endress, 1999; Alford, 2003, 2007). 2000). (8E) Scyphostegiaceae + Salicaceae (100/1). Androecial initi- (10F) Podostemaceae + Hypericaceae (100/1). Stamens in pha- ation centrifugal in polystemonous groups of Salicaceae. langes (in Podostemaceae a single phalanx). Ovules with ex- (Gilg, 1925; van Heel, 1967; Bernhard and Endress, 1999). tremely thin nucelli, tendency of the outer integument to be only 2 cell layers thick. Podostemaceae are highly specialized (9) Violaceae + Goupiaceae (75/0.62). Flowers haplostemo- water plants with reduced flowers. Thus, the two families are nous. Anthers with conspicuous apical appendage. Nectary, morphologically very different. (Rao, 1957; Arekal and if present, at outer base of stamens. Ovules more than 2 per Nagendran, 1977; Rutishauser, 1997). carpel, antitropous, crassinucellar (ovule structure in Goupiaceae unknown). (Melchior and Becker, 1925; (11) Pandoids (Pandaceae + Irvingiaceae) (64/0.97). Loesener, 1942; Singh, 1963). Placentation axile. Ovule 1 per carpel, antitropous, crassinu- cellar. (Merino Sutter et al., 2006; Matthews and Endress, (10) Clusioids + ochnoids (70/0.81). Petals often contort. 2011; Tobe and Raven, 2011). Mostly polystemonous. Placentation mostly axile. Nucelli often thin (ovules often incompletely tenuinucellar), with (12) Malpighioids (Centroplacaceae + Malpighiaceae + endothelium, outer integument tends to be thicker than inner Elatinaceae) (63/0.51). Low stamen number (haplostemonous Downloaded from (most clusioides, part of ochnoids). (Matthews et al., 2012). or diplostemonous). Placentation axile. Sepals persistent in . (Gilg, 1908; Niedenzu, 1925; Loesener, 1942; (10A) Ochnoids (Ochnaceae + Medusagynaceae + Quiinaceae) Anderson, 1990). (100/1). Sepals often of very different size within a flower; sepals with more than 3 vascular traces. Petals not delayed (12A) Malpighiaceae + Elatinaceae (100/1). Sepals with secre- in floral bud development but forming the protective organs tory structures on dorsal side. Mostly diplostemonous. http://aob.oxfordjournals.org/ of advanced ; petal aestivation predominantly contort; Carpels mostly 3; tendency to have dorsally bulged ovaries; petals reflexed at anthesis, with 3 vascular traces. placentation axile. Ovules without endothelium, tendency of Androecium mostly polystemonous (diplostemonous or hap- outer integument to be only 2 cell layers thick. Presence of lostemonous in some Ochnaceae); anthers basifixed; anthers cleistogamous flowers in some representatives of both fam- x-shaped; short androgynophore and gynophore. Tendency to ilies. However, it should also be mentioned that the two fam- more than 5 carpels per flower; stigmas more or less ilies markedly differ in showy vs. minute, inconspicuous suction-cup-shaped; vasculature forming a dorsal band of flowers, monosymmetry vs. polysymmetry, uniovular vs. plur-

bundles in the upper stylar region; micropyle formed by both iovular carpels, crassinucellar vs. weakly crassinucellar ovules at Universitaet Zuerich on April 20, 2013 integuments; gynoecium epidermis with large, radially elong- (Niedenzu, 1925; Frisendahl, 1927; Raghavan and Srinivasan, ate cells. Flowers nectarless. Special mucilage cells in floral 1940; Rao, 1940; Dathan and Singh, 1971; Anderson, 1980, organs; tanniferous tissue and sclerenchyma abundant in 1990; Vogel, 1990; Davis and Chase, 2004; Davis and floral organs. (Dickison, 1990b; Amaral, 1991; Matthews Anderson, 2010; W.-H. Zhang et al., 2012). et al., 2012). (10B) Medusagynaceae + Quiinaceae (75/0.98). Flowers (at ASSOCIATIONS OF FAMILIES BASED MERELY least partly) functionally and morphologically unisexual. ON FLORAL STRUCTURAL SIMILARITIES (NOT Polystemonous. Massive thecal septum that persists after de- FORMING CLADES IN MOLECULAR STUDIES) hiscence. Very similar gynoecium with more than 5 carpels, stigmatic lobes radiating outward from the ovary, stigmas capi- There also appear to be groups of families, which share struc- tate (suction-cup-shaped), large ovary roof, ovary surface con- tural similarities, that are not always closely related according spicuously sculpted with longitudinal ribs. Ovules 2 (or more) to recent molecular phylogenetic studies. In fact, in some cases per carpel, superposed, with a ‘false endothelium’ on the nu- they appear quite distantly related. It is of interest to address cellus surface. (Matthews et al., 2012). these groups because they may tell us something about parallel evolutionary trends within Malpighiales or of future relation- (10C) Clusioids (Bonnetiaceae + Clusiaceae + Calophyl- ships yet to be resolved. Numbers in parentheses refer to the laceae + Podostemaceae + Hypericaceae) (100/1). Stamens three main clades and the subclade numbers in the preceding often in phalanges (groups of united stamens). Placentation section, to which the families belong. axile. Nucelli always thin, outer integument thicker than inner (not in Hypericaceae). (Engler, 1925; Rao, 1957; Prakash and Lau, 1976; Arekal and Nagendran, 1977; Lim, Malesherbiaceae, Passifloraceae, Turneraceae (1: 8A), Violaceae 1984; Rutishauser, 1997; Moura˜o and Beltrati, 2000; (1: 9). Five stamens, 3 carpels, parietal placenta, corona, aril, Sweeney, 2008). crassinucellar ovules, several ovules per carpel. Aril lacking in Malesherbiaceae (Gengler-Nowak, 2003), but chalazal (10D) Clusiaceae + Bonnetiaceae (85/0.92). The 5 (or 4) sepals appendage present (Harms, 1925a), as in ; corona are preceded by several . (Engler, 1925; Dickison and lacking in Violaceae. Thin integuments: Malesherbiaceae, Weitzman, 1998; Sweeney, 2008). Passifloraceae, Turneraceae (outer 2, inner 3), Violaceae (10E) Calophyllaceae + Podostemaceae + Hypericaceae (100/ (both integuments 3) (Singh, 1963; Ricardi, 1967; Kloos and 1). Difficult to characterize. The only studied species of Bouman, 1980). Androgynophore in Malesherbiaceae and Calophyllaceae (in Mammea) has unitegmic ovules (in contrast Passifloraceae (not in Turneraceae and Violaceae). Pollen 980 Endress et al. — Floral structure and systematics in Malpighiales with more than three colporate apertures and thick two-layered A CASE STUDY: EVOLUTIONARY CHANGES IN or channelled apertural intine in Passifloraceae and Violaceae OVULES IN MALPIGHIALES (Furness, 2011). We have demonstrated here that features of the ovules repre- sent a rich source of macroevolutionary information. A con- Euphorbiaceae (1: 4B), Malpighiaceae (3: 12A). Anderson spicuous feature in many Malpighiales are ovules with thin, (1990) considered Euphorbiaceae and Malpighiaceae as close slender nucelli and an endothelium. They are either weakly possible relatives. In Soltis et al. (2011) Malpighiaceae plus crassinucellar or even incompletely tenuinucellar (Fig. 1). If Elatinaceae appear as sister to Phyllanthaceae plus we first focus on the entire COM clade, we see that this Picrodendraceae but Euphorbiaceae are far apart. This alterna- feature is not restricted to Malpighiales but is common in tive topology, however, is not well supported. Secretory struc- closely related Celastrales ( and , tures on the dorsal surface and secretory fimbriae on margins Matthews and Endress, 2005a) and also in a number of of sepals, 3 carpels, tendency to dorsally bulged ovaries, Oxalidales [Cephalotaceae, some , some axile placenta (the combination of 3 carpels and axile placenta s.l. (); Matthews and Endress, is unusual at the level of Malpighiales), a single ovule per 2002)]. Thus, it is probably not a synapomorphy for just carpel, ovules crassinucellar, at least partly campylotropous, Malpighiales; it may, however, be a synapomorphy for the nucellar beak protruding out of at least the inner integument, entire COM clade. The feature dominates in clade 2 Downloaded from gaps between inner integument and nucellus and between (clusioids + ochnoids + rhizophoroids + pandoids; but does the integuments, lack of an endothelium; occasional presence not occur in Ctenolophonaceae and Irvingiaceae, and is of several meiocytes in the ovules. Penaea type of embryo unknown in Pandaceae) and in clade 3 (chrysobalanoids + sac development. (Modilewski, 1911; Rao, 1941; Singh, malpighioids + putranjivoids; but does not occur in

1965; Sutter and Endress, 1995). Zigzag micropyles occasion- Malpighiaceae and Caryocaraceae, and is unknown in http://aob.oxfordjournals.org/ ally present (e.g. Bouharmont, 1962; Singh, 1959). Centroplacaceae). Another interesting exception are the Malpighian hairs (Euphorbiaceae, , Correll and derived mangrove genera of Rhizophoraceae (e.g. Correll, 1982; see also Pax and Hoffmann, 1931). Presence Rhizophora) with most likely secondarily crassinucellar of latex (Vega et al., 2002). ovules because of their large seeds concomitant with their spe- cialized fruit biology. Conversely, thin nucelli occur as excep- Linaceae (1: 6), Humiriaceae (1: 7A), Erythroxylaceae, tions in Rafflesiaceae and some Linaceae among clade 4 Rhizophoraceae (2: 2B), Ochnaceae (2: 10A), Chrysobalanaceae, (euphorbioids), which otherwise have crassinucellar ovules. Trigoniaceae (3: 3C), Lophopyxidaceae (3: 3G). These families at Universitaet Zuerich on April 20, 2013 show tendencies to diplostemony, stamens basally united, nec- BIG GAPS IN OUR KNOWLEDGE taries at outer stamen base (or at inner petal base), axile pla- centa, two collateral antitropous ovules per carpel, with In 13 families of Malpighiales development of ovules and em- obturator, ovules incompletely tenuinucellar or weakly crassi- bryology are largely unknown, such as in Caryocaraceae, nucellar, with endothelium. This may be a plesiomorphic com- Centroplacaceae, Euphroniaceae, Goupiaceae, Ixonanthaceae, bination of features. Lacistemataceae, Lophopyxidaceae, Malesherbiaceae, Medusagynaceae, Pandaceae, Peraceae and Quiinaceae (Table 1). In most families there are at least some rudimentary Podostemaceae, Hypericaceae (2: 10F), Elatinaceae (3: 12A). Water plants (among Hypericaceae, elodes), data on internal floral morphology and floral anatomy. But many ovules per carpel, placenta only basally axile (in even these are unstudied in Centroplacaceae, Goupiaceae and Lacistemataceae (Fig. 1). These are obvious areas where Podostemaceae entirely axile), nucellus conspicuously elong- future inquiry would be fruitful. ating at its base in Elatinaceae and Podostemaceae.

Achariaceae (1: 7C), Samydaceae (1: 8D), Scyphostegiaceae, CONCLUSIONS current Salicaceae (1: 8E) (i.e. former Flacourtiaceae pro The distribution of floral characters in Malpighiales shows parte). Lacerate arils, constriction of the funicle in that, as expected, variation of features is at different systematic and Salix, Populus; embryo sac growing out of levels. For instance, fusion or non-fusion of sepals is variable ovules in , , and Populus; oligoste- within families (e.g. Euphorbiaceae and Ochnaceae), whereas mony combined with synandry in Scyphostegia and some nucellus thickness is constant in families and even in suprafa- Salix species. milial clades (e.g. thick in Euphorbiaceae and the parietal clade, and thin in clusioids, ochnoids and chrysobalanoids; Euphorbiaceae (1: 4B), Phyllanthaceae, Picrodendraceae (1: Fig. 1). In this paper we have focused on features that are rela- 5A). Unisexual flowers. Petals mostly lacking. Ovules antitro- tively constant within families but vary among families, pro- pous, with obturator, crassinucellar, with prominent nucellar viding information on suprafamilial clades. Such features cap, sometimes nucellar beak (ovules unknown in Peraceae) mainly occur in gynoecium and ovule structure. Notably, pla- (Kapil and Bhatnagar, 1994; Sutter and Endress, 1995; centation form, direction of ovule curvature in combination Merino Sutter et al., 2006), aril differentiated as a caruncle with presence or absence of an obturator, thickness of the nu- (for Picrodendraceae, Berg, 1975), fruits explosive, falling cellus and relative thickness of the two integuments are fea- into pieces. tures of special interest. Endress et al. — Floral structure and systematics in Malpighiales 981

Our original studies in several subclades of Malpighiales Amaral MCE, Bittrich V. 1998. Ontogenia inicial do androceu de espe´cies de identified many new potentially significant features for fam- Ochnaceae subfam. Sauvagesioideae atrave´s de ana´lise em microscopia eletroˆnica de varredura. Revista Brasileira de Botaˆnica 21: 269–273. ilies of the order, especially in gynoecium and ovule structure. Ameka KG, Pfeifer E, Rutishauser R. 2002. Developmental morphology of Together with the literature these studies have begun to reveal amicorum and S. submersa (Podostemaceae: Podostemoideae) patterns of structural characteristics for the newly recognized from Ghana. Botanical Journal of the Linnean Society 139: 255–273. major suprafamilial subclades in recent molecular studies. Anderson C. 2007. Revision of (Malpighiaceae). Contribution They also begin to give us an idea of how evolution may from the University of Michigan Herbarium 25: 1–82. have proceeded within the large clade Malpighiales. As in Anderson WR. 1980. Cryptic self-fertilization in the Malpighiaceae. Science 207: 892–893. other organisms, complexes of characters can disappear but re- Anderson WR. 1990. The origin of the Malpighiaceae – The evidence from appear because the genetic structure for the complex has not morphology. Memoirs of the New York Botanical Garden 64: 210–224. been lost but rather persists in a cryptic state so that the char- Anderson WR. 2006. Eight segregates from the neotropical genus acter or character complex can easily reappear under certain (Malpighiaceae). Novon 16: 168–204. circumstances. This is expressed by trends of evolution and APG I (The Angiosperm Phylogeny Group). 1998. An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical parallel evolution. We are still only at the beginning of under- Garden 85: 531–553. standing extreme evolutionary pathways, such as in euphor- APG III. 2009. An update of the Angiosperm Phylogeny Group classification bioids with the advent of the giant flowers of Rafflesiaceae for the orders and families of flowering plants: APG III. Botanical Downloaded from and the tiny reduced flowers of Euphorbia (Davis et al., Journal of the Linnean Society 161: 105–121. 2007, 2008), or in Malpighiaceae with the evolution of floral Arbo MM. 2007. Turneraceae. In Kubitzki K. ed. The families and genera of monosymmetry (W.-H. Zhang et al., 2012). vascular plants, 9. Berlin: Springer, 458–466. Arekal GD, Nagendran CR. 1977. The female gametophyte in two Indian Thus, a number of subclades of Malpighiales newly found genera of Tristichoideae (Podostemaceae) – a reinvestigation. in molecular phylogenetic studies can also be recognized as Proceedings of the Indian Academy of Sciences B 86: 287–294. related by shared floral structural features. Nevertheless, Armbruster WS. 1984. The role of resin in angiosperm pollination: ecologic- http://aob.oxfordjournals.org/ there are still big lacunae in our structural knowledge of the al and chemical considerations. American Journal of Botany 71: other subclades at the suprafamilial level, and a lack of reso- 1149–1160. Baas P. 1970. Anatomical contributions to plant I. Floral and vege- lution in parts of the Malpighiales phylogeny. Knowledge of tative anatomy of Eliaea from Madagascar and from the structure of the families and larger subclades of Indo-Malesia (Guttiferae). Blumea 18: 369–391. Malpighiales is important for reconstruction of the evolution Baas P. 1972. Anatomical contributions to II. The affinities of of the group and a better understanding of its biological prop- Hua Pierre and Afrostyrax Perkins et Gilg. Blumea 20: 369–391. erties, and also for the placement of fossil flowers (Endress and Baum-Leinfellner H. 1953. U¨ ber unifaziale Griffel und Narben. Planta 42: 452–460.

Friis, 2006; Scho¨nenberger and von Balthazar, 2006; Friis at Universitaet Zuerich on April 20, 2013 Bayer C. 2007. Huaceae. In Kubitzki K. ed. The families and genera of vas- et al., 2011). The future looks bright! cular plants, 9. Berlin: Springer, 191–193. Berg R. 1975. Fruit, seed, and myrmecochorous dispersal in (Euphorbiaceae). Norwegian Journal of Botany 22: 173–194. Bernhard A. 1999a. Floral structure and development of Ceratiosicyos laevis ACKNOWLEDGEMENTS (Achariaceae) and its systematic position. Botanical Journal of the Linnean Society 131: 103–113. This study is part of the project ‘ diversity and evolution Bernhard A. 1999b. Flower structure, development, and systematics in in rosids’ of P.K.E., supported by grant # 31003A_129804 of Passifloraceae and in (Flacourtiaceae). International Journal of the Swiss National Foundation (SNF). In addition, studies for Plant Sciences 160: 135–150. the phylogenetic component for this research was funded by Bernhard A, Endress PK. 1999. Androecial development and systematics in the United States National Science Foundation (NSF) Flacourtiaceae s.l. Plant Systematics and Evolution 215: 141–155. Assembling the Tree of Life Grant DEB-0622764 (to Bigio NC, Secco RS. 2012. As espe´cies de (Euphorbiaceae s.s.)in Brazilian Amazon. Rodrigue´sia 63 (1): 163–207. C.C.D.), and NSF DEB-1120243 (to C.C.D.). For helpful in- Bittrich V, Amaral MCE. 1996. Pollination biology of globulifera formation on floral structure in Malpighiales we thank Maria (Clusiaceae). Plant Systematics and Evolution 200: 101–110. do Carmo E. Amaral, Volker Bittrich and Rolf Rutishauser. Bittrich V, Amaral MCE. 1997. Floral biology of some Clusia species from For comments on the manuscript we thank Joel Nitta and Central Amazonia. Kew Bulletin 52: 617–635. Zhenxiang Xi. For critical reviews of the manuscript we Bittrich V, Amaral MCE, Machado SMF, Marsaioli AJ. 2003. Floral resin of Tovomitopsis saldanhae (Guttiferae) and 7-Epi-nemorosone: structural thank Louis Ronse De Craene and an anonymous reviewer. revision. Zeitschrift fu¨r Naturforschung 58c: 643–648. Boesewinkel FD. 1985. The ovule and seed of balsamifera (Aubl.) St. Hil. Acta Botanica Neerlandica 34: 183–191. Boesewinkel FD, Bouman F. 1980. Development of ovule and seed-coat of LITERATURE CITED mombuttense Engl. with notes on the species. Acta Agostini G. 1973. El genero Mutis (Lacistemaceae). Acta Botanica Botanica Neerlandica 29: 103–115. Venezuelica 8: 167–175. Bor J, Bouman F. 1975. Development of ovule and integuments in Euphorbia Airy Shaw HK. 1974. Cleistanthus insignis Airy Shaw. Hooker’s Icones milii and . Phytomorphology 24: 280–296. Plantarum 38 (1): Tabula 3711. Bouharmont J. 1962. Fe´condation de l’ovule et de´veloppement de la graine Alford MH. 2003. Claves para los ge´neros de Flacourtiaceae de ´ y del apre`s croisement et autopollinisation chez brasiliensis. La Nuevo Mundo. Arnaldoa 10: 19–38. Cellule 62: 119–130. Alford MH. 2007. Samydaceae. Tolweb.org/Samydaceae/68361/2007.02.06 Bouman F, Meijer W. 1994. Comparative structure of ovules and seeds in Alford MH. 2008. Revision of (Salicaceae). Systematic Botany Rafflesiaceae. Plant Systematics and Evolution 193: 187–212. Monographs 85: 1–62. Chase MW, Soltis DE, Olmstead RG, et al. 1993. of seed Amaral MCE. 1991. Phylogenetische Systematik der Ochnaceae. Botanische plants: an analysis of nucleotide sequences from the plastid gene rbcL. Jahrbu¨cher fu¨r Systematik 113: 105–195. Annals of the Missouri Botanical Garden 80: 528–580. 982 Endress et al. — Floral structure and systematics in Malpighiales

Chase MW, Zmarzty S, Lledo´ MD, Wurdack KJ, Swensen SM, Fay MF. Engler A. 1925. Guttiferae. In Engler A, Prantl K. eds. Die natu¨rlichen 2002. When in doubt, put it in Flacourtiaceae: a molecular phylogenetic Pflanzenfamilien, 2nd edn, 21. Leipzig: Engelmann, 154–237. analysis based on plastid DNA sequences. Kew Bulletin 57: 141–181. Finet C, Timme RE, Delwiche CF, Marle´taz F. 2010. Multigene phylogeny Corner EJH. 1976. The seeds of . Cambridge: Cambridge of the green reveals the origin and diversification of land plants. University Press. Current Biology 20: 2217–2222. Correll DS, Correll HB. 1982. Flora of the Bahama archipelago. Vaduz: Friis EM, Crane PR, Pedersen KR. 2011. Early flowers and angiosperm evo- Cramer. lution. Cambridge: Cambridge University Press. Cronquist A. 1981. An integrated system of classification of flowering plants. Frisendahl A. 1927. U¨ ber die Entwicklung chasmogamer und kleistogamer New York: Columbia University Press. Blu¨ten bei der Gattung . Acta Horti Goetheborgensis 3: 99–142. Cuatrecasas J. 1961. A taxonomic revision of the Humiriaceae. Contributions Furness CA. 2011. Comparative structure and development of pollen and from the United States National Herbarium 35: 25–214. tapetum in Malpighiales, with a focus on the parietal clade. Dathan ASR, Singh D. 1971. Embryology and seed development in L. International Journal of Plant Sciences 172: 980–1011. Journal of the Indian Botanical Society 50: 362–370. Ganders FR. 1979. Heterostyly in (Erythroxylaceae). Dathan ASR, Singh D. 1979. Structure and development of female gameto- Botanical Journal of the Linnean Society 78: 11–20. phyte and seed in laurifolia (Dennst.) Sleumer. Journal Gengler-Nowak KM. 2003. Molecular phylogeny and taxonomy of of the Indian Botanical Society 58: 256–263. Malesherbiaceae. Systematic Botany 28: 333–344. Davis CC, Anderson WR. 2010. A complete generic phylogeny of Gilg E. 1908. Flacourtiaceae africanae. Botanische Jahrbu¨cher fu¨r Systematik Malpighiaceae inferred from nucleotide sequence data and morphology. 40: 444–518.

American Journal of Botany 97: 2031–2048. Gilg E. 1925. Flacourtiaceae. In Engler A, Prantl K. eds. Die natu¨rlichen Downloaded from Davis CC, Chase MW. 2004. Elatinaceae are sister to Malpighiaceae; Pflanzenfamilien, 2nd edn, 25. Leipzig: Engelmann, 377–457. Peridiscaceae belong to . American Journal of Botany 91: Gonza´lez AM. 1993. Anatomı´a y vascularizacio´n floral de race- 262–273. mosa, hassleriana y Turnera joelii (Turneraceae). Bonplandia Davis CC, Webb CO, Wurdack KJ, Jaramillo CA, Donoghue MJ. 2005. 7: 143–184. Explosive radiation of Malpighiales supports a mid- origin Gustafsson MHG. 2000. Floral morphology and relationshipsof Clusia gun- of modern tropical rain . American Naturalist 165: E36–E65. dlachii with a discussion of floral organ identity and diversity in the

Davis CC, Latvis M, Nickrent DL, Wurdack KJ, Baum DA. 2007. Floral genus Clusia. International Journal of Plant Sciences 161: 43–53. http://aob.oxfordjournals.org/ gigantism in Rafflesiaceae. Science 315: 1812. Gustafsson MHG, Bittrich V. 2002. Evolution of morphological diversity and Davis CC, Endress PK, Baum DA. 2008. Evolution of floral gigantism. resin in flowers of Clusia (Clusiaceae): insights from ITS se- Current Opinion in Plant Sciences 11: 49–57. quence variation. Nordic Journal of Botany 22: 183–203. De-Paula OC, Sajo MG, Prenner G, Cordeiro I, Rudall PJ. 2011. Harms H. 1925a. Malesherbiaceae. In Engler A, Prantl K. eds. Die natu¨rli- Morphology, development and homologies of the perianth and floral nec- chen Pflanzenfamilien, 2nd edn, 21. Leipzig: Engelmann, 467–470. taries in Croton and (Euphorbiaceae-Malpighiales). Plant Harms H. 1925b. Passifloraceae. In Engler A, Prantl K. eds. Die natu¨rlichen Systematics and Evolution 292: 1–14. Pflanzenfamilien, 2nd edn, 21. Leipzig: Engelmann, 470–507. Dehgan B, Webster GL. 1979. Morphology and infrageneric relationships of Hemingway CA, Christensen AR, Malcomber ST. 2011. B- and C-class the genus (Euphorbiaceae). University of California gene expression during corona development of the blue passionflower

Publications, Botany, 74: 1–73. (Passiflora caerulea, Passifloraceae). American Journal of Botany 98: at Universitaet Zuerich on April 20, 2013 Dickison WC. 1990a. A study of the floral morphology and anatomy of the 923–934. Caryocaraceae. Bulletin of the Torrey Botanical Club 117: 123–137. Hilu KW, Borsch T, Mu¨ller K, et al. 2003. Angiosperm phylogeny based on Dickison WC. 1990b. The morphology and relationships of matK sequence information. American Journal of Botany 90: 1758–1776. (Medusagynaceae). Plant Systematics and Evolution 171: 27–55. Hochwallner H, Weber A. 2006. Flower development and anatomy of Clusia Dickison WC, Weitzman AL. 1998. Floral morphology and anatomy of valerioi, a Central American species of Clusiaceae offering floral resin. Bonnetiaceae. Journal of the Torrey Botanical Society 125: 268–286. Flora 201: 407–418. Doyle JA, Endress PK. 2000. Morphological phylogenetic analysis of basal Igersheim A, Endress PK. 1998. Gynoecium diversity and systematics of the angiosperms: comparison and combination with molecular data. paleoherbs. Botanical Journal of the Linnean Society 127: 289–370. International Journal of Plant Sciences 161: S121–S153. Ja¨ger-Zu¨rn I. 1967. Embryologische Untersuchungen an vier Podostemaceen. Endress PK. 1994. Diversity and evolutionary biology of tropical flowers. O¨ sterreichische Botanische Zeitschrift 114: 20–45. Cambridge: Cambridge University Press. Ja¨ger-Zu¨rn I. 2003. The occurrence of apical septum in the ovary of Endress PK. 1999. Symmetry in flowers – diversity and evolution. , , , and Mourera (Podostemoideae- International Journal of Plant Sciences 160: S3–S23. Podostemaceae): a taxonomic implication. Botanische Jahrbu¨cher fu¨r Endress PK. 2010. Flower structure and trends of evolution in eudicots and Systematik 124: 303–324. their major subclades. Annals of the Missouri Botanical Garden 97: Juel O. 1915. U¨ ber den Bau des Gynoeceums bei Parinarium. Arkiv fo¨r 541–583. Botanik 14: 1–12. Endress PK. 2011a. Angiosperm ovules: diversity, development, evolution. Kamelina OP, Konnova VA. 1990. Embryological characters of Annals of Botany 107: 1465–1489. Biebersteinia. Dokladhoi Akademijai Fanhoi RSDS Togikis 33: 193–195. Endress PK. 2011b. Evolutionary diversification of the flowers in angios- Kapil RN. 1956. A further contribution to the morphology and life history of perms. American Journal of Botany 98: 370–396. Chrozophora Neck. Phytomorphology 6: 278–288. Endress PK. 2012. The immense diversity of floral monosymmetry and asym- Kapil RN. 1970. Podostemaceae. Bulletin of the Indian National Science metry across angiosperms. Botanical Review 78: 345–397. Academy 41: 104–109. Endress PK, Friis EM. 2006. Rosids – reproductive structure, fossil and Kapil RN, Bhatnagar AK. 1994. The contribution of embryology to the sys- extant, and their bearing on deep relationships. Introduction. Plant tematics of Euphorbiaceae. Annals of the Missouri Botanical Garden 81: Systematics and Evolution 260: 83–85. 145–159. Endress PK, Igersheim A. 2000. Gynoecium structure and evolution in basal Kapil RN, Tiwari SC. 1978. The integumentary tapetum. Botanical Review angiosperms. International Journal of Plant Sciences 161: S211–S223. 44: 457–490. Endress PK, Matthews ML. 2006a. Elaborate petals and in eudi- Kaul RB. 1995. Reproductive structure and organogenesis in a cottonwood, cots: structure, , evolution. 6 Organisms, Diversity and Evolution (Salicaceae). International Journal of Plant Sciences 6: 257–293. 156: 172–180. Endress PK, Matthews ML. 2006b. First steps towards a floral structural Kloos A, Bouman F. 1980. Case studies in aril development. Passiflora sub- characterization of the major rosid subclades. Plant Systematics and erosa L. and L. Beitra¨ge zur Biologie der Pflanzen 55: Evolution 260: 223–251. 49–66. Endress PK, Matthews ML. 2012. Progress and problems in the assessment Knapp S, Mallet J. 1984. Two new species of Passiflora (Passifloraceae), of flower morphology in higher-level systematics. Plant Systematics and with comments on their natural history. Annals of the Missouri Evolution 298: 257–276. Botanical Garden 71: 1068–1074. Endress et al. — Floral structure and systematics in Malpighiales 983

Korotkova N, Schneider JV, Quandt D, Worberg A, Zizka G, Borsch T. Mauritzon J. 1934. Etwas u¨ber die Embryologie der Zygophyllaceen sowie 2009. Phylogeny of the eudicot order Malpighiales: analysis of a recalci- einige Fragmente u¨ber die der Humiriaceen. Botaniska Notiser 1934: trant clade with sequences of petD group II intron. Plant Systematics and 409–422. Evolution 282: 201–228. Mauritzon J. 1936. Zur Embryologie einiger Parietales-Familien. Svensk Krause K. 1925. Lacistemaceae. In Engler A, Prantl K. eds. Die natu¨rlichen Botanisk Tidskrift 30: 79–113. Pflanzenfamilien, 2nd edn, 21. Leipzig: Engelmann, 321–323. Melchior H. 1964. A. Engler’s Syllabus der Pflanzenfamilien, 12th edn, 2. Krosnick SE, Harris EM, Freudenstein JV. 2006. Patterns of anomalous Berlin-Nikolassee: Borntraeger. floral development in the Asian Passiflora (subgenus Decaloba; supersec- Melchior H, Becker W. 1925. Violaceae. In Engler A, Prantl K. eds. Die nat- tion Disemma). American Journal of Botany 93: 620–636. u¨rlichen Pflanzenfamilien, 2nd edn, 21. Leipzig: Engelmann, 329–377. Kubitzki K. 2007a. Introduction to the Passifloraceae alliance Merino Sutter D, Endress PK. 2003. Structure of female flowers and cupules (“Passiflorales”¼Malpighiales). In Kubitzki K. ed. The families and in Balanopaceae, an enigmatic rosid family. Annals of Botany 92: genera of vascular plants, 9. Berlin: Springer, 12. 459–469. Kubitzki K. 2007b. Malesherbiaceae. In Kubitzki K. ed. The families and Merino Sutter D, Forster PI, Endress PK. 2006. Female flowers and system- genera of vascular plants, 9. Berlin: Springer, 247–249. atic position of Picrodendraceae (Euphorbiaceae s.l., Malpighiales). Plant Lee EK, Cibrian-Jaramillo A, Kolokotronis SO, et al. 2011. A functional Systematics and Evolution 261: 187–215. phylogenomic view of the seed plants. PLoS Genetics 7: e1002411. Michaelis P. 1924. Blu¨tenmorphologische Untersuchungen an den http://dx.doi.org/10.1371/journal.pgen.1002411. Euphorbiaceen unter besonderer Beru¨cksichtigung der Phylogenie der Leins P. 1964. Die fru¨he Blu¨tenentwicklung von Angiospermenblu¨te. Botanische Abhandlungen 3: 1–150. Wight et Arn. und H. aegypticum L. Berichte der Deutschen Mildbraed J. 1931. Pandaceae. In Engler A, Prantl K. eds. Die natu¨rlichen Botanischen Gesellschaft 77: 112–123. Pflanzenfamilien, 2nd edn, 19a. Leipzig: Engelmann, 1–3. Downloaded from Leins P, Erbar C. 1991. Fascicled androecia in Dilleniidae and some remarks Mitchell JD. 2002. Goupiaceae. In Mori SA, Cremers G, Gracie CA, de on the androecium. Botanica Acta 104: 336–344. Granville J-J, Heald SV, Hoff M, Mitchell JD. eds. Guide to the vascular Lemke DE. 1988. A synopsis of Flacourtiaceae. Aliso 12: 29–43. plants of Central French Guiana. Memoirs of the New York Botanical Lim AL. 1984. The embryology of Garcinia mangostana L. (Clusiaceae). Garden 76: 199–202. Gardens’ Bulletin Singapore 37: 93–103. Modilewski J. 1910. Weitere Beitra¨ge zur Embryobildung einiger

Link DA. 1992. The floral nectaries of the and their systematic Euphorbiaceen. Berichte der Deutschen Botanischen Gesellschaft 28: http://aob.oxfordjournals.org/ implications: VI. Ixonanthaceae Exell and Mendonc¸a. Botanische 413–418. Jahrbu¨cher fu¨r Systematik 114: 81–90. Modilewski J. 1911. U¨ ber die anomale Embryosackentwicklung bei Litt A, Chase MW. 1998. The systematic position of , with com- L. und anderen Euphorbiaceen. Berichte der ments on the position of : an analysis based on rbcL sequence Deutschen Botanischen Gesellschaft 29: 430–436. data. Systematic Botany 23: 401–409. Moline P, Thiv M, Ameka GK, Ghogue J-P, Pfeifer E, Rutishauser R. Lloyd DG, Webb CJ. 1992. The evolution of heterostyly. In Barrett SCH. ed. 2007. Comparative morphology and molecular systematics of African Evolution and function of heterostyly. Berlin: Springer, 151–207. Podostemaceae-Podostemoideae with emphasis on Dicraeanthus and Loesener T. 1942. Celastraceae. In Engler A, Prantl K. eds. Die natu¨rlichen from Cameroon. International Journal of Plant Pflanzenfamilien, 2nd edn, 20b. Leipzig: Engelmann, 87–197. Sciences 168: 159–180. Losos JB. 2011. Seeing the forest for the trees: the limitation of phylogenies in Morton CM. 2011. Newly sequenced nuclear gene (Xdh) for inferring angio- comparative biology. American Naturalist 177: 709–727. sperm phylogeny. Annals of the Missouri Botanical Garden 98: 63–89. at Universitaet Zuerich on April 20, 2013 Matthews ML, Endress PK. 2002. Comparative floral structure and systema- Moura˜o KSM, Beltrati CM. 2000. Morphology and anatomy of developing tics in Oxalidales (Oxalidaceae, , Cephalotaceae, fruits and seeds of L. (Clusiaceae). Revista Brunelliaceae, , Elaeocarpaceae, Tremandraceae). Brasileira de Biologia 60: 701–711. Botanical Journal of the Linnean Society 140: 321–381. Mukherjee PK. 1958. The female gametophyte of Acalypha malabarica Matthews ML, Endress PK. 2004. Comparative floral structure and systema- Muell. with a brief discussion on the Penaea type of embryo-sac. tics in Cucurbitales (Corynocarpaceae, Coriariaceae, Datiscaceae, Journal of the Indian Botanical Society 37: 504–508. , , , ). Nandi OI, Chase MW, Endress PK. 1998. A combined cladistic analysis of Botanical Journal of the Linnean Society 145: 129–185. angiosperms unsing rbcL and non-molecular data sets. Annals of the Matthews ML, Endress PK. 2005a. Comparative floral structure and systema- Missouri Botanical Garden 85: 137–212. tics in Celastrales (Celastraceae, Parnassiaceae, Lepidobotryaceae). Narayana LL. 1970. Linaceae. Bulletin of the Indian National Science Botanical Journal of the Linnean Society 149: 129–194. Academy 41: 127–132. Matthews ML, Endress PK. 2005b. Comparative floral structure and systema- Narayana LL, Rao D. 1969a. Contributions to the floral anatomy of tics in Crossosomatales (Crossosomataceae, Stachyuraceae, Staphyleaceae, Humiriaceae 1. Journal of Japanese Botany 44: 328–335. Aphloiaceae, Geissolomataceae, Ixerbaceae, ). Botanical Narayana LL, Rao D. 1969b. Contributions to the floral anatomy of Linaceae Journal of the Linnean Society 147: 1–46. (1). Journal of Japanese Botany 44: 289–294. Matthews ML, Endress PK. 2006. Floral structure and systematics in four Narayanaswami S, Sawhney S. 1959. Microsporogenesis and embryo sac de- orders of rosids, including a broad survey of floral mucilage cells. velopment in Casearia tomentosa Roxb. Phyton (Buenos Aires) 13: Plant Systematics and Evolution 260: 199–221. 133–144. Matthews ML, Endress PK. 2008. Comparative floral structure and systema- Niedenzu F. 1925. Elatinaceae. In Engler A, Prantl K. eds. Die natu¨rlichen tics in Chrysobalanaceae s.l. (Chrysobalanaceae, Dichapetalaceae, Pflanzenfamilien, 2nd edn, 21. Leipzig: Engelmann, 270–276. Euphroniaceae, Trigoniaceae; Malpighiales). Botanical Journal of the Nikiticheva ZI, Yakovlev MS. 1985. Rhizophoraceae. In Batygina TB, Linnean Society 157: 249–309. Yakovlev MS. eds. Comparative embryology of flowering plants – Matthews ML, Endress PK. 2011. Comparative floral structure and systema- Brunelliaceae-Tremandraceae. Leningrad: Nauka, 120–125. tics in Rhizophoraceae, Erythroxylaceae and the potentially related Nikolayeva ES. 1983. Salicaceae. In Yakovlev MS. ed. Comparative embry- Ctenolophonaceae, Linaceae, Irvingiaceae and Caryocaraceae ology of flowering plants – Phytolaccaceae-. Leningrad: (Malpighiales). Botanical Journal of the Linnean Society 166: 331–416. Nauka, 188–192. Matthews ML, Endress PK. 2013. Comparative floral structure and systema- Qiu YL, Li L, Wang B, et al. 2010. Angiosperm phylogeny inferred from tics of the clade of Lophopyxidaceae and Putranjivaceae. Botanical sequences of four mitochondrial genes. Journal of Systematics and Journal of the Linnean Society, in press. Evolution 48: 391–425. Matthews ML, Endress PK, Scho¨nenberger J, Friis EM. 2001. A compari- Pauze´ F, Sattler R. 1978. L’androce´es centripe`te d’Ochna atropurpurea. son of floral structures of Anisophylleaceae and Cunoniaceae and the Canadian Journal of Botany 56: 2500–2511. problem of their systematic position. Annals of Botany 88: 439–455. Pax F, Hoffmann K. 1931. Euphorbiaceae. In Engler A, Prantl K. eds. Die Matthews ML, Amaral MEC, Endress PK. 2012. Comparative floral struc- natu¨rlichen Pflanzenfamilien, 2nd edn, 19c. Leipzig: Engelmann, 11– ture and systematics in Ochnaceae s.l. (Ochnaceae, Quiinaceae, and 233, 240. Medusagynaceae; Malpighiales). Botanical Journal of the Linnean Philbrick CT, Novelo M. 2004. Monograph of Podostemum (Podostemaceae). Society 170: 299–392. Systematic Botany Monographs 70: 1–106. 984 Endress et al. — Floral structure and systematics in Malpighiales

Prakash N, Lau YY. 1976. Morphology of alternifolium and the Schwarzbach AE, Ricklefs RE. 2000. Systematic affinities of taxonomic position of Ploiarium. Botaniska Notiser 129: 279–285. Rhizophoraceae, based on DNA, nuclear ribsomal DNA, Prance GT, da Silva MF. 1973. Caryocaraceae. Flora Neotropica Monograph and morphology. American Journal of Botany 87: 547–564. 12: 1–77. Schweiger J. 1905. Beitra¨ge zur Kenntnis der Samenentwicklung der Prance GT, White F. 1988. The genera of Chrysobalanaceae – A study in Euphorbiaceen. Flora 94: 339–379. practical and theoretical taxonomy and its relevance to evolutionary Shulaev V, Sargent DJ, Crowhurst RN, et al. 2010. The genome of - biology. Philosophical Transactions of the Royal Society London B land strawberry (Fragaria vesca). Nature Genetics 43: 109–116. 320: 1–184. Simpson BB. 1989. Pollination biology and taxonomy of and Prenner G, Rudall PJ. 2007. Comparative ontogeny of the cyathium in Dinemagonum (Malpighiaceae). Systematic Botany 14: 408–426. Euphorbia and its allies: exploring the organ-flower-inflorescence bound- Singh B. 1959. Studies in the family Malpighiaceae: I. Morphology of ary. American Journal of Botany 94: 1612–1629. glauca Kuntze. Horticultural Advance (Saharanpur) 3: 1–19. Prenner G, Hopper SD, Rudall PJ. 2008a. Pseudanthium development in Singh D. 1963. Structure and development of ovule and seed of paucifolius, with particular reference to the evolution of L. and Ionidium suffruticosum Ging. Journal of the Indian Botanical the cyathium in (Euphorbiaceae-Malpighiales). Australian Society 42: 448–462. Systematic Botany 21: 153–161. Singh RP. 1965. Structure and development of seeds in Codiaeum variegatum Prenner G, Box MWS, Cunniff J, Rudall PJ. 2008b. The branching stamens Blume. Journal of the Indian Botanical Society 44: 205–210. of Ricinus and the homologies of the angiosperm stamen fascicle. Sleumer H. 1954. Flacourtiaceae. In van Steenis CGGJ. ed. Flora Malesiana, International Journal of Plant Sciences 169: 735–744. Ser I, 5. Leiden: Noordhoff, 1–106.

Radcliffe-Smith A. 2001. Genera Euphorbiacearum. Kew: Royal Botanic Sobral-Leite M, Siqueira Filho JA, Erbar C, Machado IC. 2011. Downloaded from Gardens Kew. Anthecology and reproductive system of Mourera fluviatilis Raghavan TS, Srinivasan VK. 1940. A contribution to the life-history of (Podostemaceae): pollination by and xenogamy in a predominantly Bergia capensis Linn. Journal of the Indian Botanical Society 19: anemophilous and autogamous family? Aquatic Botany 95: 77–87. 283–291. Soltis DE, Soltis PS, Chase MW, et al. 2000. Angiosperm phylogeny inferred Rao AMS. 1940. Studies in the Malpighiaceae. 1. Embryo-sac development from 18S rDNA, rbcL, and atpB sequences. Botanical Journal of the and embryogeny in the genera , Banisteria and Stigmaphyllum. Linnean Society 133: 381–461.

Journal of the Indian Botanical Society 18: 145–156. Soltis DE, Smith SA, Cellinese N, et al. 2011. Angiosperm phylogeny: 17 http://aob.oxfordjournals.org/ Rao AMS. 1941. Studies in the Malpighiaceae. 2. Structure and development genes, 640 taxa. American Journal of Botany 98: 704–730. of the ovules and embryo-sacs of Malpighia coccifera Linn. and Souto LS, Oliveira DMT. 2008. Morfoanatomia e ontogeˆnese das sementes Tristellateia australis Linn. Proceedings of the National Institute of de espe´cies de C.B. Robinson e A. Juss. Science B 7: 393–404. (Malpighiaceae). Acta Botanica Brasilica 22: 733–740. Rao AN. 1957. The embryology of Thunb. and Steinmann VW. 2005. New Euphorbiaceae from . II. Contributions H. mysorense Heyne. Phytomorphology 7: 36–45. from the University of Michigan Herbarium 24: 173–187. Ricardi SM. 1967. Revision taxonomica de las Malesherbiaceas. Gayana Bot. Stenar H. 1937. Zur Embryosackentiwicklung einiger Malpighiazeen. 16: 1–139. Botaniska Notiser 1937: 110–118. Robson NKB. 1996. Studies in the genus Hypericum Stephens EL. 1909. The embryo-sac and embryo of certain .

L. (Guttiferae). 6. Sections 20. Myriandra to 28. Elodes. Bulletin of the Annals of Botany 23: 363–378. at Universitaet Zuerich on April 20, 2013 Natural History Museum 26: 75–217. Stevens PF. 1976. The Old World species of (Guttiferae). I. The Robson NKB. 2012. Studies in the genus Hypericum Mascarene species. Journal of the Arnold Arboretum 57: 167–184. L. (Hypericaceae). 9. Addenda, corrigenda, keys, lists and general discus- Stevens PF. 2001 onwards. Angiosperm phylogeny website, version 12, July sion. 72: 1–111. 2012 [and more or less continuously updated since]. Available at http:// Ronse Decraene LP, Smets E. 1991. Androecium and floral nectaries of www.mobot.org/MOBOT/research/APweb/ madagascariensis. Plant Systematics and Evolution 178: Stevens PF. 2007. Clusiaceae-Guttiferae. In Kubitzki K. ed. The families and 179–194. genera of vascular plants, 9. Berlin: Springer, 48–66. Ronse Decraene LP, Smets E. 1992. Complex polyandry in the Magnoliatae: Steyermark JA, Luteyn JL. 1980. Revision of the genus definition, distribution and systematic value. Nordic Journal of Botany (Linaceae). Brittonia 32: 128–143. 12: 621–649. Steyn EMA, van Wyk AE, Smith GF. 2003. Embryology and systematic rela- Ruhfel BR, Bittrich V, Bove CP, et al. 2011. Phylogeny of the clusioid clade tionships of (Flacourtiaceae). Bothalia 33: 199–206. (Malpighiales): evidence from the plastid and mitochondrial genomes. Steyn EMA, Smith GF, van Wyk AE. 2004. Functional and taxonomic sig- American Journal of Botany 98: 306–325. nificance of seed structure in (Salicaceae). Bothalia 34: Rutishauser R. 1997. Structural and developmental diversity in 53–59. Podostemaceae (river-weeds). Aquatic Botany 57: 29–70. Steyn EMA, van Wyk AE, Smith GF. 2005. Ovule and seed structure in Rutishauser R, Grubert M. 1999. The architecture of Mourera fluviatilis zeyheri (Scolopieae) with notes on the embryology of (Podostemaceae): developmental morphology of inflorescences, flowers, Salicaceae. Bothalia 35: 175–183. and seedlings. American Journal of Botany 86: 907–922. Sutter D, Endress PK. 1995. Aspects of gynoecium structure and macrosyste- Sabatier D. 2002. ovicarpa (Humiriaceae), a new species from matics in Euphorbiaceae. Botanische Jahrbu¨cher fu¨r Systematik 116: French Guiana. Brittonia 54: 233–235. 517–536. Sabatier D. 2004. Humiriaceae. In Smith N, Mori SA, Henderson A, Sweeney PW. 2008. Phylogeny and floral diversity in the genus Garcinia Stevenson DW, Heald SV. eds. Flowering plants of the Neotropics. (Clusiaceae) and relatives. International Journal of Plant Sciences 169: Princeton: Princeton University Press, 185–187. 1288–1303. Sacchi CF, Price PW. 1988. Pollination of the arroyo , : Takhtajan A. 1997. Diversity and classification of flowering plants. role of insects and wind. American Journal of Botany 75: 1387–1393. New York: Columbia University Press. Sastre C. 1970. Recherches sur les Ochnace´es. II. Les espe`ces de Sauvagesia Tateishi S. 1927. On the development of the embryo sac and fertilization of L. a` placenta basale. Caldasia 10: 497–516. . Botanical Magazine Tokyo 41: 477–485. Savolainen V, Chase MW, Hoot SB, et al. 2000a. Phylogenetics of flowering Thiv M, Ghogue J-P, Grob V, Huber K, Pfeifer E, Rutishauser R. 2009. plants based on combined analysis of plastid atpB and rbcL gene How to get off the mismatch at the generic rank in African sequences. Systematic Biology 49: 306–362. Podostemaceae. Plant Systematics and Evolution 283: 57–77. Savolainen V, Fay MF, Albach DC, et al. 2000b. Phylogeny of the eudicots: a Tobe H, Raven PH. 1984. An embryological contribution to systematics of nearly complete familial analysis based on rbcL gene sequences. Kew the Chrysobalanaceae I. Chrysobalaneae. Botanical Magazine Bulletin 55: 257–309. Tokyo 97: 397–411. Scho¨nenberger J, von Balthazar M. 2006. Reproductive structures and Tobe H, Raven PH. 2011. Embryology of the Irvingiaceae, a family with un- phylogenetic framework of the rosids – progress and prospects. Plant certain relationships among the Malpighiales. Journal of Plant Research Systematics and Evolution 260: 87–106. 124: 577–591. Endress et al. — Floral structure and systematics in Malpighiales 985

Tokuoka T, Tobe H. 1995. Embryology and systematics of Euphorbiaceae Webster GL. 1994a. Classification of the Euphorbiaceae. Annals of the sens. lat.: a review and perspective. Journal of Plant Research 108: Missouri Botanical Garden 81: 3–32. 97–106. Webster GL. 1994b. Synopsis of the genera and suprageneric taxa of Tokuoka T, Tobe H. 1998. Ovules and seeds in Crotonoideae Euphorbiaceae. Annals of the Missouri Botanical Garden 81: 33–144. (Euphorbiaceae): structure and systematic implications. Botanische Winkler H. 1931. Linaceae. In Engler A, Prantl K. eds. Die natu¨rlichen Jahrbu¨cher fu¨r Systematik 120: 165–186. Pflanzenfamilien, 2nd edn, 19a. Leipzig: Engelmann, 82–130. Tokuoka T, Tobe H. 1999. Embryology of tribe Drypeteae, an enigmatic Wurdack KJ, Davis CC. 2009. Malpighiales phylogenetics: gaining ground taxon of Euphorbiaceae. Plant Systematics and Evolution 215: 189–208. on one of the most recalcitrant clades in the angiosperm tree of life. Tokuoka T, Tobe H. 2001. Ovules and seeds in subfamily Phyllanthoideae American Journal of Botany 96: 1551–1570. (Euphorbiaceae): structure and systematic implications. Journal of Plant Xi Z-X, Ruhfel BR, Schaefer H, et al. 2012. Phylogenomics and a posteriori Research 114: 75–92. data partitioning resolve the Cretaceous angiosperm radiation Tokuoka T, Tobe H. 2002. Ovules and seeds in Malpighiales. Proceedings of the National Academy of Sciences of the (Euphorbiaceae): structure and systematic implications. Journal of Plant USA 109: 17519–17524. Research 115: 361–374. Zhang L-B, Simmons MP. 2006. Phylogeny and delimitation of the Tokuoka T, Tobe H. 2003. Ovules and seeds in Acalyphoideae Celastrales inferred from nuclear and plastid genes. Systematic Botany (Euphorbiaceae): structure and systematic implications. Journal of Plant 31: 122–137. Research 116: 355–380. Tokuoka T, Tobe H. 2006. Phylogenetic analyses of Malpighiales using Zhang N, Zeng LP, Shan HY, Ma H. 2012. Highly conserved low-copy plastid and nuclear DNA sequences, with particular reference to the em- nuclear genes as effective markers for phylogenetic analyses in angios- bryology of Euphorbiaceae sens. str. Journal of Plant Research 119: perms. New Phytologist 195: 923–937. Downloaded from 599–616. Zhang W-H, Kramer EM, Davis CC. 2012. Similar genetic mechanisms van Heel WA. 1967. Anatomical and ontogenetic investigations on the morph- underlie the parallel evolution of floral phenotypes. PLoS one 7: ology of the flowers and the fruit of Scyphostegia borneensis Stapf e36033. http://dx.doi.org/10.1371/journal.pone.0036033. (Scyphostegiaceae). Blumea 15: 107–125. Zhang Z-G, Meng A-P, Li J-Q, Ye Q-G, Wang H-C, Endress PK. 2012. Vega AS, Castro MA, Anderson WR. 2002. Occurrence and phylogenetic Floral development of chekiangensis (Phyllanthaceae), with significance of latex in the Malpighiaceae. American Journal of Botany special reference to androecium and gynoecium. Plant Systematics and http://aob.oxfordjournals.org/ 89: 1725–1729. Evolution 298: 1229–1238. Vogel S. 1974. O¨ lblumen und o¨lsammelnde Bienen. Tropische und Zhu XY, Chase MW, Qiu YL, Kong HZ, Dilcher D, Chen ZD. 2007. Subtropische Pflanzenwelt 7: 283–547. Mitochondrial matR sequences help to resolve deep phylogenetic relation- Vogel S. 1990. History of the Malpighiaceae in the light of pollination ships in rosids. BMC Evolutionary Biology 7: 217. http://dx.doi.org/ ecology. Memoirs of the New York Botanical Garden 55: 130–142. 10.1186/1471-2148-7-217. at Universitaet Zuerich on April 20, 2013